Hoffbrand’s Essential Haematology Ninth Edition A. Victor Hoffbrand MA DM FRCP FRCPath FRCP(Edin) DSc FMedSci Emeritus Professor of Haematology University College London Contributing Authors Pratima Chowdary MBBS MD MRCP FRCPath Professor of Haemophilia and Haemostasis, University College London And Consultant Haematologist KD Haemophilia and Thrombosis Centre Royal Free Hospital, London, UK Graham P. Collins MA MBBS FRCP FRCPath DPhil Associate Professor of Haematology and Haematology Consultant Oxford Cancer and Haematology Centre Churchill Hospital, Oxford, UK Justin Loke BM BCh PhD MRCP FRCPath AACR-CRUK Transatlantic Fellow Dana-Farber Cancer Institute, Boston, USA and University of Birmingham, UK
Contents Preface to the Ninth Edition iv About the Companion Website v 1 Haemopoiesis 1 2 Erythropoiesis and general aspects of anaemia 11 3 Hypochromic anaemias 27 4 Iron overload 42 5 Megaloblastic anaemias and other macrocytic anaemias 50 6 Haemolytic anaemias 65 7 Genetic disorders of haemoglobin 79 8 The white cells, part 1: granulocytes, monocytes and their benign disorders 99 9 The white cells, part 2: lymphocytes and their benign disorders 116 10 The spleen 131 11 The aetiology and genetics of haematological neoplasia 138 12 Management of haematological malignancy 155 13 Acute myeloid leukaemia 168 14 Chronic myeloid leukaemia 185 15 Myeloproliferative neoplasms 194 16 Myelodysplastic neoplasms 210 17 Acute lymphoblastic leukaemia 222 18 The chronic lymphocytic leukaemias 236 19 Hodgkin lymphoma 246 20 Non-Hodgkin lymphomas 1: General aspects 259 21 Non-Hodgkin lymphomas 2: Individual diseases 270 22 Multiple myeloma and related plasma cell neoplasms 287 23 Amyloid 302 24 Aplastic anaemia and bone marrow failure syndromes 309 25 Haemopoietic stem cell transplantation 320 26 Platelets, coagulation and normal haemostasis 337 27 Bleeding disorders caused by platelet and vascular abnormalities 355 28 Hereditary coagulation disorders 370 29 Acquired coagulation disorders and thrombotic microangiopathies 385 30 Thrombosis 1: Pathogenesis and diagnosis 399 31 Thrombosis 2: Treatment 417 32 Haematological changes in systemic diseases 437 33 Blood transfusion 451 34 Pregnancy and neonatal haematology 465 Appendix: 5 th edition (2022) of the World Health Organization Classification of Haematolymphoid Tumours 475 Index 478
Preface to the Ninth Edition Advances in the understanding the pathogenesis of blood diseases and improvements in their treatment have continued apace in the 5 years since the eighth edition of Hoffbrand’s Essential Haematology was published. Gene mutations are increasingly used to define and classify inherited and acquired haematological diseases, as a guide to therapy and to predict prognosis. Mutations underlying many rarer blood diseases have been identified, allowing appropriate panels of DNA probes to be estab- lished, facilitating the diagnosis of future cases. Many more drugs directed against specific sites in the cell signalling pathways have been approved. The past five years have also seen substantial advances in immunological treatment for malignant haematological diseases. Mono- and bi-specific antibodies are increasingly incorporated into frontline therapy as well into treatment of relapsed disease. Chimeric antigen receptor (CAR)-T cells are challenging stem cell transplantation for potential cure for relapsed B-cell lym- phoid neoplasms. New drugs have also been introduced for treatment of benign (now termed in the United States ‘classical’) haematological diseases. These include mitapivat for pyruvate kinase deficiency, sutimlimab for cold agglutinin disease, luspatercept for anaemia in thalassaemia and myelodysplasia and pegcetacoplan for paroxysmal nocturnal haemoglobinuria. Drugs inhibiting prolyl hydroxylase in the hypoxia-inducible factor pathway are being developed to treat anaemia. They are already in illegal use for ‘dop- ing’ of athletes to enhance their performance. The fifth edition of the World Health Organisation (2022) Classification of the Haemato-lymphoid Tumours has been incorporated throughout this new edition and is given as an Appendix. The International Consensus Classification (ICC) of Myeloid Neoplasms, Acute Leukaemias and Mature Lymphoid Neoplasms was also published in 2022. It is beyond the scope of this book to compare and contrast the WHO and ICC classifications. Reference to the ICC classification are given in the Appendix. David Steensma, the remarkable co-author of HEH8 stepped down for this new edition when he was appointed Global Head of Haematology at Novartis Institute for Biological Research. For the first time for Essential Haematology, there will be co- authorship by a specialist in the coagulation field. Professor Pratima Chowdary, Director of the Katherine Dormandy Haemophilia and Thrombosis Centre at the Royal Free Hospital, London, has ensured that the major section of the book dealing with bleeding and thrombotic disorders is authoritative and up to date. Graham Collins, Associate Professor of Haematology, Oxford Haematology and Cancer Centre has had the monumental task of updating the sections of HEH dealing with the lymphoid malignancies. Dr Justin Loke, AACR-CRUK Transatlantic Fellow, University of Birmingham, UK, now at the Dana-Farber Cancer Institute, Boston, USA, has undertaken the parallel task of updating the chapters dealing with the myeloid malignancies. We are grateful to Dr Connor Sweeney, Professor Ashutosh Wechelaker and Professor Irene Roberts for their expert contri- butions to chapters 17 (acute lymphoblastic leukaemia), 23 (amyloid) and 34 (pregnancy and neonatal haematology), respec- tively. We are also grateful to Professor Barbara Bain who kindly checked the validity of our accompanying MCQs and to Dr Kirollos Kamel for his valuable contributions to chapters 30 and 31 (thrombosis and its management). We thank our publishers Wiley-Blackwell and especially Sophie Bradwell, Mandy Collison, Neelukiran Sekar and Kimberly Monroe-Hill for their unstinting help and support at all stages of production of this new edition. We also thank Jane Fallows for producing again such beautiful, clear diagrams. Essential Haematology began life in 1980 as a textbook for medical students. We hope medical and other undergraduate students will continue to use it and share our excitement about one of the most advanced fields in medicine. With the vast expan- sion of knowledge of blood and its diseases over the past 44 years, the book has inevitably expanded. It is now also suitable for those beginning a career in haematology, for clinical and non-clinical scientists and nurses with an interest in blood and its diseases and for those working in closely related fields. A. Victor Hoffbrand London, 2024
About the Companion Website Don’t forget to visit the companion website for this book: www.wiley.com/go/haematology9e ere you will find invaluable material designed to enhance your learning, including: Multiple Choice Questions Figures (PPT) Tables (PDF)
2 / Chapter 1: Haemopoiesis This chapter deals with the general aspects of blood cell forma- tion (haemopoiesis). The processes that regulate haemopoiesis and the early stages of formation of red cells (erythropoiesis), granulocytes and monocytes (myelopoiesis) and platelets (thrombopoiesis) are also discussed. Site of haemopoiesis In the first few weeks of gestation, the embryonic yolk sac is a transient site of primitive haemopoiesis. Definitive haemopoie- sis derives from a population of stem cells first observed in the aorta–gonads–mesonephros (AGM) region of the developing embryo. These common precursors of endothelial and hae- mopoietic cells are called haemangioblasts and seed the liver, spleen and bone marrow. From 6 weeks until 6–7 months of foetal life, the liver and spleen are the major haemopoietic organs and continue to pro- duce blood cells until about 2 weeks after birth (Table 1.1; see Fig. 7.1b). The placenta also contributes to foetal haemopoie- sis. The bone marrow takes over as the most important site from 6 to 7 months of foetal life. During normal childhood and adult life, the marrow is the only source of new red cells, granulocytes, monocytes and platelets. The developing cells are situated outside the bone marrow sinuses; mature cells are released into the sinus spaces, the marrow microcirculation and so into the general circulation. In infancy all the bone marrow is haemopoietic, but during childhood and beyond there is progressive replace- ment of marrow throughout the long bones with fat cells, so that in adult life haemopoietic marrow is confined to the central skeleton and proximal ends of the femurs and humeri (Table1.1). Even in these active haemopoietic areas, approximately 50% of the marrow consists of fat in the middle-aged adult (Fig.1.1). The remaining fatty mar- row is capable of reversion to haemopoiesis, and in many dis- eases there is also expansion of haemopoiesis down the long bones. Moreover, in certain disease states, the liver and spleen can resume their foetal haemopoietic role (‘extramedullary haemopoiesis’). Haemopoietic stem and progenitor cells Haemopoiesis starts with a pluripotent stem cell that can self- renew by asymmetrical cell division but also gives rise to the precursor of the separate cell lineages. The stem cells are able to repopulate a bone marrow from which all stem cells have been eliminated by lethal irradiation or chemotherapy. Self- renewal and repopulating ability define the haemopoietic stem cell (HSC). HSCs are rare perhaps 1 in every 20 million nucleated cells in bone marrow. Newer DNA sequencing techniques suggest that a typical adult has approximately 50 000 HSCs. HSCs are heterogeneous, with some able to repopulate a bone marrow for more than 16 weeks, called long-term HSCs, while others, although able to produce all haemopoi- etic cell types, engraft only transiently for a few weeks and are called short-term HSCs . Although the exact cell surface marker phenotype of the HSC is still unknown, on immunological testing these cells are positive for the markers cluster of differentiation 34 (CD34), CD49f and CD90 and negative for CD38 and CD45RA and for cell lineage-defining markers (Lin). Morphologically, HSCs have the appearance of small- or medium-sized lymphocytes. Cell differentiation occurs from the stem cells via commit- ted haemopoietic progenitors, which are restricted in their developmental potential (Fig. 1.2). The existence of the separate progenitor cells can be demonstrated by in vitro culture tech- niques. Stem cells and very early progenitors are assayed by cul- ture on bone marrow stroma as long-term culture-initiating cells, whereas later progenitors are generally assayed in semi- solid media. As examples, in the erythroid series progenitors can be identified in special cultures as burst-forming units (BFU-E, describing the ‘burst’ with which they form in culture) and Table 1.1 Dominant sites of haemopoiesis at different stages of development. Foetus 0–2 months (yolk sac) 2–7 months (liver, spleen) 5–9 months (bone marrow) Infants Bone marrow (practically all bones); dwindling contribution from liver/spleen that ceases in the first few months of life Adults Vertebrae, ribs, sternum, skull, sacrum and pelvis, proximal ends of femur Figure 1.1 Normal bone marrow trephine biopsy (posterior iliac crest). Haematoxylin and eosin stain; approximately 50% of the intertrabecular tissue is haemopoietic tissue and 50% fat.
Chapter 1: Haemopoiesis / 3 colony-forming units (CFU-E; Fig 1.2); the mixed granulocyte/ monocyte progenitor is identified as a colony-forming unit- granulocyte/monocyte (CFU-GM) in culture. Megakaryocytes derive from a megakaryocyte progenitor, itself derived from an earlier mixed erythroid–megakaryocyte progenitor. In the haemopoietic hierarchy, the pluripotent stem cell gives rise to a mixed erythroid and megakaryocyte progeni- tor, which then divides into separate erythroid and megakary- ocyte progenitors. The pluripotent stem cell also gives rise to a mixed lymphoid, granulocyte and monocyte progenitor, which divides into a progenitor of granulocytes and monocytes and a mixed lymphoid progenitor, from which B- and T-cell lymphocytes and natural killer (NK) cells develop (Fig. 1.2). The spleen, lymph nodes and thymus are secondary sites of lymphocyte production (Chapter 9). As the stem cell has the capability for self-renewal (Fig. 1.3), the marrow cellularity remains constant in a normal, healthy steady state. There is considerable amplification in the system: one stem cell is capable of producing about 10 6 mature blood cells after 20 cell divisions (Fig. 1.3). In humans, HSCs are capable of about 50 cell divisions (the ‘Hayflick limit’), with progressive telomere shortening with each division affect- ing viability. Under normal conditions most HSCs are dormant, with at most only a few percent active in cell cycle on any given day. Any given HSC enters the cell cycle approximately once every 3 months to 3 years in humans. By contrast, progenitor cells are much more numerous and highly proliferative. With ageing, the number of stem cells falls and the relative propor- tion giving rise to lymphoid rather than myeloid progenitors also falls. Stem cells also accumulate genetic mutations with age, an average of 8 exonic coding mutations by age 60 years (1.3 per decade). These, either passengers without oncogenic potential or drivers that cause clonal expansion, may be present in neoplasms arising from these stem cells (Chapters 11, 16). The progenitor and precursor cells are capable of respond- ing to haemopoietic growth factors with increased production of one or other cell line when the need arises. The development BFU E , burst-forming unit erythroid CFU-E, colony-forming unit erythroid Pluripotent stem cell Megakaryocyte erythroid progenitor BFU E Megakaryocyte progenitor Granulocyte monocyte progenitor Monocyte progenitor Granulocyte progenitor Neutrophil progenitor Eosinophil progenitor Basophil, mast cell progenitor Thymus Lymphoid myeloid progenitor Multipotential lymphoid progenitor Erythroid progenitors CFU-E Red cells Platelets Monocytes Neutrophils Eosinophils Basophils Lymphocytes Natural killer cells B T NK Figure 1.2 Diagrammatic representation of the bone marrow pluripotent stem cells (haemopoietic stem cells, HSC) and the cell lines that arise from them. A megakaryocytic/erythroid progenitor (MkEP) and a mixed lymphoid/myeloid progenitor are formed from the pluripotent stem cells. Each gives rise to more differentiated progenitors. BFU-E, burst-forming unit erythroid; CFU-E, colony-forming unit erythroid.
4 / Chapter 1: Haemopoiesis of the mature cells (red cells, granulocytes, monocytes, mega- karyocytes and lymphocytes) is considered further in other sec- tions of this book. Bone marrow stroma and niches The bone marrow forms a suitable environment for stem cell survival, self-renewal and formation of differentiated progeni- tor cells. It is composed of various types of stromal cells and a microvascular network (Fig. 1.4). The stromal cells include adipocytes, fibroblasts, macrophages, megakaryocytes, osteoblasts, osteoclasts, endothelial cells and mesenchymal stem cells (which have the capacity to self-renew and dif- ferentiate into osteocytes, adipocytes and chondrocytes). The stromal cells secrete extracellular molecules such as colla- gen, glycoproteins (fibronectin and thrombospondin) and gly- cosaminoglycans (hyaluronic acid and chondroitin derivatives) to form an extracellular matrix. The HSCs reside in two types of niche. These provide some of the growth factors, adhesion molecules and cytokines which support stem cells, maintaining their viability and reproduc- tion, e.g. stem cell factor (SCF) expressed by stromal and endothelial cells binds to its receptor, KIT (CD117), on stem cells. The niches are either vascular, including arterioles and sinusoids that converge on a central vein, or endosteal with osteoblasts and osteoclasts closely associated with bone. Sympathetic nerves and non-myelinated Schwann cells are important regulators of stem cell quiescence or release. Haemopoietic stem cells (as well as mesenchymal stem cells) traffic around the body. They are found in peripheral blood in low numbers. In order to exit the bone marrow, cells must cross the blood vessel endothelium, and this process of mobilization is enhanced for HSCs by the administration of growth factors such as granulocyte colony-stimulating factor (G-CSF). The reverse process, stem cell homing, depends on a chemokine gradient in which stromal-derived factor 1 (SDF-1), which binds to its receptor CXCR4 on HSC, is critical. Stem cell Extracellular matrix Fibroblast Adhesion molecule Growth factor Ligand Growth factor receptor Endothelial cell or osteoblast Macrophage Mesenchymal stem cell Fat cell Figure 1.4 Haemopoiesis occurs in a suitable microenvironment (‘niche’) provided by a stromal matrix on which stem cells grow and divide. The niche may be vascular (lined by endothelium) or endosteal (lined by osteoblasts). There are specific recognition and adhesion sites; extracellular glycoproteins, e.g. fibronectin, collagen and other compounds, form a matrix and are involved in stem cell binding (see text). (b) Mature cells Recognizable committed marrow precursors Multipotent progenitor cells Stem cells Self renewal or Quiescent Proliferation Differentiation (a) Figure 1.3 (a) Bone marrow cells are increasingly differentiated and lose the capacity for self-renewal as they mature. (b) A single stem cell gives rise, after multiple cell divisions (shown by vertical lines), to >10 6 mature cells.
Chapter 1: Haemopoiesis / 5 The regulation of haemopoiesis Transcription factors Haemopoiesis starts with stem cell division in which one cell replaces the stem cell (self-renewal) and the other is committed to differentiation. These early committed pro- genitors express low levels of transcription factors that com- mit them to discrete cell lineages. Transcription factors regulate gene expression by control- ling the transcription of specific genes or gene families (Fig. 1.5). Typically, they contain at least two domains: a DNA-binding domain, such as a leucine zipper or helix–loop– helix motif which binds to a specific DNA sequence, and an activation domain, which contributes to the assembly of the transcription complex at a gene promoter. The transcription factors interact, so that reinforcement of one transcription pro- gramme may suppress that of another lineage Which cell lineage is selected for differentiation depends on both chance and the external signals received by progenitor cells. Examples of transcription factors involved in haemopoie- sis include RUNX1, GATA2 and MT2A in the earliest stages; GATA1, GATA2 and FOG1 in erythropoiesis and megakaryo- cytic differentiation; PU.1 and the CEBP family in granu- lopoiesis; PAX5 in B lymphocyte and NOTCH1 in T lymphocyte development. The transcription factors induce synthesis of proteins specific to a cell lineage. For example, GATA1 binds to specific motifs on the erythroid genes for glo- bin and haem synthesis and so activates these genes. Mutation, deletion or translocation of transcription factor genes underlies many cases of haematological neoplasms (Chapter 11). Haemopoietic growth factors The haemopoietic growth factors are a group of glycoproteins that regulate the proliferation and differentiation of haemopoi- etic progenitor cells and the function of mature blood cells. They may act locally at the site where they are produced by cell–cell contact, e.g. SCF, or circulate in plasma, e.g. G-CSF or erythropoietin (EPO). They also bind to the extracellular matrix to form niches to which stem and progenitor cells adhere. The growth factors may cause cell proliferation, but can also stimulate differentiation and maturation, prevent apoptosis and affect the function of mature cells (Fig. 1.6). The growth factors share a number of common properties (Table 1.2) and act at different stages of haemopoiesis (Table 1.3; Fig. 1.6). Stromal cells are the major source of growth factors except for EPO, 90% of which is synthesized in the kidney, and thrombopoietin (TPO), made largely in RNA polymerase + accessory factors Transcription DNA-binding domain Transactivation domain Enhancer DNA sequence TATA box sequence (promotor) Structural gene Figure 1.5 Model for control of gene expression by a transcription factor. The DNA-binding domain of a transcription factor binds a specific enhancer sequence adjacent to a structural gene. The transactivation domain then binds a molecule of RNA polymerase, thus augmenting its binding to the TATA box. The RNA polymerase now initiates transcription of the structural gene to form mRNA. Translation of the mRNA by the ribosomes generates the protein encoded by the gene. Transcription factors work in combination to both activate and repress the expression of a large number of genes. Activation of phagocytosis, killing, secretion Proliferation Differentiation Maturation Early cell Late cell G-CSF G-CSF G-CSF Suppression of apoptosis Functional activation G-CSF Monocyte Neutrophil G-CSF Figure 1.6 Growth factors may stimulate the proliferation of early bone marrow cells, direct differentiation to one or other cell type, stimulate cell maturation, suppress apoptosis or affect the function of mature non-dividing cells, as illustrated here for granulocyte colony-stimulating factor (G-CSF) for an early myeloid progenitor and a mature neutrophil.
6 / Chapter 1: Haemopoiesis the liver. An important feature of growth factor action is that two or more factors may synergize in stimulating a particular cell to proliferate or differentiate. Moreover, the action of one growth factor on a cell may stimulate production of another growth factor or growth factor receptor. SCF, TPO.NOTCH1 and FLT3 ligand act locally on the pluripotential stem cells and on myeloid/lymphoid progeni- tors (Fig. 1.7). Interleukin-3 (IL-3) has widespread activity on lymphoid/myeloid and megakaryocyte/erythroid progeni- tors. Granulocyte–macrophage colony-stimulating factor (GM-CSF), G-CSF and macrophage colony-stimulating fac- tor (M-CSF) enhance neutrophil and macrophage/monocyte production, IL-5 eosinophil, KIT mast cell, TPO platelet and EPO red cell production. These lineage-specific growth fac- tors also enhance the effects of SCF, FLT3-L and IL-3 on the survival and differentiation of early haemopoietic cells. Interleukin-7 is involved at all stages of lymphocyte produc- tion, and various other interleukins and toll-like receptor ligands (not shown) direct B and T lymphocyte and NK cell production (Fig. 1.7). These factors maintain a pool of haemopoietic stem and progenitor cells on which later-acting factors, EPO, G-CSF, M-CSF, IL-5 and TPO, act to increase production of one or other cell lineage in response to the body’s need. Granulocyte and monocyte formation, for example, can be stimulated by infection or inflammation through release of IL-1 and tumour necrosis factor (TNF), which then stimulate stromal cells to produce growth factors in an interacting network (Fig. 8.4). In contrast, cytokines, such as transforming growth factor- β (TGF- β) and γ-interferon (IFN- γ), can exert a negative effect on haemopoiesis and may have a role in the development of aplastic anaemia (p. 313). Growth factor receptors and signal transduction The biological effects of growth factors are mediated through specific receptors on target cells. Many receptors, such as the EPO receptor (EPO-R) and GM-CSF-R, are from the hae- mopoietin receptor superfamily which dimerize after bind- ing their ligand. Dimerization of the receptor leads to activation of a com- plex series of intracellular signal transduction pathways, of which the three major ones are the JAK/STAT (signal trans- ducer and activator of transcription) pathway, the mitogen- activated protein (MAP) kinase and the phosphatidylinositol 3-kinase (PI3K) pathways (Fig. 1.8; see also Fig 9.4, Fig 15.2). The Janus-associated kinase (JAK) proteins are a family of four tyrosine-specific protein kinases that associate with the intra- cellular domains of the growth factor receptors (Fig. 1.8). A growth factor molecule binds simultaneously to the extracel- lular domains of two or three receptor molecules, resulting in their aggregation. Receptor aggregation induces activation of the JAKs, which then phosphorylate members of the STAT family of transcription factors. This results in their dimeriza- tion and translocation from the cell cytoplasm across the nuclear membrane to the cell nucleus. Within the nucleus STAT dimers activate the transcription of specific genes. A model for the control of gene expression by a transcription fac- tor is shown in Fig. 1.5. The clinical importance of this path- way is revealed for example by the finding of an activating mutation of the JAK2 gene as a cause of polycythaemia vera and related myeloproliferative neoplasms (p. 195). JAK can also activate the MAPK pathway, which is regulated by RAS and controls proliferation. PI3 kinases phosphorylate Table 1.3 Haemopoietic growth factors (see also Fig. 1.7). Act on stromal cells IL-1, TNF Act on pluripotential stem cells SCF, TPO, FLT3-, NOTCH1 Act on multipotent lymphoid/myeloid progenitor cells IL-3, IL-7, SCF, FLT3-L, TPO, GM-CSF Act on lineage-committed progenitor cells Granulocyte/monocyte production: IL-3, GM-CSF, G-CSF, M-CSF, IL-5 (eosinophil CSF) Mast cell production: KIT-ligand Red cell production: IL-3, EPO Platelet production: IL-3, TPO Lymphocyte/NK cell production: IL-1, IL-2, IL-4, IL-7, IL-10, other ILs CSF, colony-stimulating factor; EPO, erythropoietin; FLT3-L, FLT3 ligand; G-CSF, granulocyte colony-stimulating factor; GM-CSF, granulocyte–macrophage colony-stimulating factor; IL, interleukin; M-CSF, macrophage/monocyte colony-stimulating factor; NK, natural killer; SCF, stem cell factor (also known as TAL1); TNF, tumour necrosis factor; TPO, thrombopoietin. Table 1.2 General characteristics of myeloid and lymphoid growth factors. Glycoproteins that act at very low concentrations Act hierarchically Usually produced by many cell types Usually affect more than one lineage Usually active on stem/progenitor cells and on differentiated cells Usually show synergistic or additive interactions with other growth factors Often act on the neoplastic equivalent of a normal cell Multiple actions: proliferation, differentiation, maturation, prevention of apoptosis, functional activation
Chapter 1: Haemopoiesis / 7 inositol lipids, which have a wide range of downstream effects, including activation of AKT. This results in a block of apoptosis and other actions (Figs. 1.8, 15.2). Different domains of the intracellular receptor protein may signal for the different pro- cesses e.g. proliferation or suppression of apoptosis, mediated by growth factors. A second, smaller group of growth factors, including SCF, FLT3L and M-CSF (Table 1.3), bind to receptors that have an extracellular immunoglobulin-like domain linked via a transmembrane bridge to a cytoplasmic tyrosine kinase domain. Growth factor binding results in dimerization of these receptors and consequent activation of the tyrosine kinase domain. Phosphorylation of tyrosine residues in the receptor itself generates binding sites for signalling proteins which initiate complex cascades of biochemical events, result- ing in changes in gene expression, cell proliferation and pre- vention of apoptosis. Adhesion molecules Cell adhesion molecules (CAMs) are glycoprotein molecules which mediate the attachment of cells to each other, to the extracellular matrix and play roles in cell-cell synapse formation. They typically are composed of three domains: intracellular, transmembrane and extracellular. They are divided into four large families: integrins, immunoglobulin super family, selec- tins and cadherins. They function as ‘molecular glue’ maintain- ing tissue structure and function. The integrins are particularly Pluripotent haemopoietic stem cells MLP GMP EryP MkP SCF/IL-3 TPO EPO KIT-L mast cell basophil monocyte dendritic cell neutrophil eosinophil GM-CSF G-CSF GM-CSF M-CSF IL-4 IL-7 IL-7 IL-5 IL-15 IL-1 IL-2 IL-4 IL-7 IL-10 NK cell LMPP SCF/TPO IL-3/FLT3-L/SCF FLT3-L ST-HSC LT-HSC T cell B cell IL-7 IL-7 red cells platelets IL-3/TPO/SCF MkEP Figure 1.7 The role of growth factors in normal haemopoiesis. Multiple growth factors act on the earlier marrow stem and progenitor cells. EPO, erythropoietin; EryP, erythroid progenitor; FLT3-L, FLT3 ligand; G-CSF, granulocyte colony-stimulating factor; GM-CSF, granulocyte– macrophage colony-stimulating factor; GMP, granulocyte–macrophage progenitor; HSC, haemopoietic stem cells; IL, interleukin; LMPP, lymphoid-primed multipotential progenitor; LT, long-term; M-CSF, macrophage/monocyte colony-stimulating factor; MkEP, megakaryocyte– erythroid progenitor; MkP, megakaryocyte progenitor; MLP, multipotential lymphoid progenitor; NK, natural killer; PSC, pluripotential stem cell; SCF, stem cell factor; ST, short-term; TLR, toll-like receptor; TPO, thrombopoietin. Source: Adapted from A.V. Hoffbrand et al. (2019) Color Atlas of Clinical Hematology: Molecular and Cellular Basis of Disease, 5th edn. Reproduced with permission of John Wiley & Sons.
8 / Chapter 1: Haemopoiesis important in linking the extracellular environment including collagen, fibronectin and fibrinogen to intracellular signalling pathways. The selectins which include E (endothelial)-selectin, L (leucocyte)-selectin and P (platelet)-selectin are particularly important in the immune system in helping white cells in traf- ficking and homing. In the bone marrow CAMs attach haemopoietic precur- sors, leucocytes and platelets to various components of the extracellular matrix, to endothelium, to other surfaces and to each other. The CAMs on the surface of leucocytes and plate- lets are termed receptors and these interact with proteins termed ligands on the surface of target cells, e.g. endothelium. The molecules are important in the development and mainte- nance of inflammatory as well as immune responses, and in platelet–vessel wall and leucocyte–vessel wall interactions. Glycoprotein IIb/IIIa, for example, is a CAM, also called inte- grin IIβ/IIIα and involved in platelet adhesion to vessel walls and to each other (Chapter 26). The pattern of expression of adhesion molecules on tumour cells may determine their mode of spread and tissue localization e.g. the pattern of metastasis of carcinoma cells to specific visceral organs or bone or of non-Hodgkin lymphoma cells into a follicu- lar or diffuse pattern. The adhesion molecules may also determine whether or not cells circulate in the bloodstream or remain fixed in tissues. They may also partly determine whether or not tumour cells are susceptible to the body’s immune defences. Attempts to treat cancer and other diseases with drugs which inhibit specific adhesion molecules have so far been unsuccessful. The cell cycle The cell division cycle, generally known simply as the cell cycle, is a complex process that lies at the heart of haemopoiesis. Dysregulation of cell proliferation is also the key to the develop- ment of malignant disease. The duration of the cell cycle is vari- able between different tissues, but the basic principles remain constant. The cycle is divided into the mitotic phase (M phase), during which the cell physically divides, and interphase, during which the chromosomes are duplicated and cell growth occurs prior to division (Fig. 1.8). The M phase is further partitioned into classical mitosis, in which nuclear division is accomplished, and cytokinesis, in which cell fission occurs. The interphase is divided into three main stages: a G 1 phase, in which the cell begins to commit to replication, an S phase, during which DNA content doubles and the chromo- somes replicate, and the G 2 phase, in which the cell organelles are copied and cytoplasmic volume is increased. If cells rest prior to division, they enter a G 0 state where they can remain for long periods of time. The number of cells at each stage of the cell cycle can be assessed by exposing cells to a chemical or radiolabel that gets incorporated into newly generated DNA. The cell cycle is controlled by two checkpoints, which act as brakes to coordinate the division process, at the end of the G 1 and G 2 phases. Two major classes of molecules control these checkpoints, cyclin-dependent protein kinases (Cdk), which phosphorylate downstream protein targets, and cyclins, which bind to Cdk and regulate their activity. An example of the importance of these systems is demonstrated by mantle cell lymphoma, which results from the constitutive activation of cyclin D1 as a result of a chromosomal translocation (p. 279). Epigenetics Epigenetics refers to changes in DNA and chromatin that affect gene expression other than those that affect DNA sequence (Fig. 16.1). Gene expression ERK 1/2 MEK 1/2 Blocked apoptosis RAF RAS Active STAT dimers Activation of gene expression STATs JAK JAK AKT PI3Kinase Plasma membrane Nucleus Growth factor M G2 G1 S Rb p53 DNA damage E2F MAP kinase MAP kinase Figure 1.8 Control of haemopoiesis by growth factors. The factors act on cells expressing the corresponding receptors. Binding of a growth factor to its receptor activates the JAK/STAT, MAPK and phosphatidyl-inositol 3-kinase (PI3K) pathways (see also Fig. 15.2), which leads to transcriptional activation of specific genes. E2F is a transcription factor needed for cell transition from G1 to S phase. E2F is inhibited by the tumour suppressor gene Rb (retinoblastoma), which can be indirectly activated by p53. The synthesis and degradation of different cyclins stimulate the cell to pass through the different phases of the cell cycle. The growth factors may also suppress apoptosis by activating AKT (protein kinase B).
Chapter 1: Haemopoiesis / 9 Cellular DNA is packaged by wrapping it around histones, a group of specialized nuclear proteins. The complex is tightly compacted as chromatin. In order for the DNA code to be read, transcription factors and other proteins need to physically attach to DNA. Histones act as custodians for this access and so for gene expression. Histones may be modified by methyla- tion, acetylation and phosphorylation, which can result in increased or decreased gene expression and so changes in cell phenotype. Epigenetics also includes changes to DNA itself, such as methylation of DNA bases. The methylation of cytosine resi- dues to methylcytosine results in inhibition of gene transcrip- tion. The DNA methyltransferase genes DNMT3A and B are involved in this methylation. TET1, 2, 3 and IDH1 and IDH2 are involved in the hydroxylation and breakdown of methylcy- tosine and restoration of gene expression (Fig. 16.1). These genes are frequently mutated in the myeloid malignancies, especially myelodysplastic syndromes and acute myeloid leu- kaemia (Chapters 13, 15 and 16). Apoptosis Apoptosis (programmed cell death) is a regulated process of physiological cell death in which individual cells are triggered to activate intracellular proteins that lead to the death of the cell. Morphologically it is characterized by cell shrinkage, con- densation of the nuclear chromatin, fragmentation of the nucleus and cleavage of DNA at inter-nucleosomal sites. It is an important process for maintaining tissue homeostasis in haemopoiesis and lymphocyte development. Apoptosis results from the action of intracellular cysteine proteases called caspases, which are activated following cleav- age and lead to endonuclease digestion of DNA and disinte- gration of the cell skeleton (Fig. 1.9). There are two major pathways by which caspases can be activated. The first is by activation through membrane proteins such as Fas or TNF receptor via their intracellular death domain. An example of this mechanism is shown by activated cytotoxic T cells express- ing Fas ligand, which induces apoptosis in target cells. The second pathway is via the release of cytochrome c from mito- chondria. Cytochrome c binds to APAF-1, which then acti- vates caspases. DNA damage induced by irradiation or chemotherapy may act through this pathway. The protein p53 encoded by the TP53 gene on chromo- some 17 has an important role in sensing DNA damage. It activates apoptosis by raising the cell level of BAX, which then increases cytochrome c release (Fig. 1.9). p53 also shuts down the cell cycle to stop the damaged cell from dividing (Fig. 1.8). The cellular level of p53 is controlled by a second protein, MDM2. Following death, apoptotic cells display molecules that lead to their ingestion by macrophages. Loss of TP53 is a major mechanism by which malignant cells evade controls that would induce cell death. As well as molecules that mediate apoptosis, there are several intracellular proteins that protect cells from apoptosis. The best-characterized example is BCL-2. BCL-2 is the proto- type of a family of related proteins, some of which are anti- apoptotic and some, like BAX, pro-apoptotic. The intracellular ratio of BAX and BCL-2 determines the relative susceptibility of cells to apoptosis, e.g. determines the lifespan of platelets, and may act through regulation of cytochrome c release from mitochondria. Many of the genetic changes associated with malignant dis- ease lead to a reduced rate of apoptosis and hence prolonged cell survival. The clearest example is the translocation of the BCL2 gene to the immunoglobulin heavy chain locus in the t(14;18) translocation in follicular lymphoma (p. xxx). Over- expression of the BCL-2 protein makes the malignant B cells less susceptible to apoptosis. The drug venetoclax which inhibits BCL-2 is now widely used to treat both myeloid and lymphoid malignant diseases. Apoptosis is the normal fate for most B cells undergoing selection in the lymphoid germinal centres. Several translocations leading to the generation of fusion pro- teins, such as t(9;22), t(11;14) and t(15;17), also result in inhibi- tion of apoptosis (Chapter 11). In addition, genes encoding APOPTOSIS BCL-2 Increased BCL-2 Survival factor e.g. growth factor Death factor e.g. Fas ligand Release of cytochrome c Inhibits Death domain Procaspases Caspases DNA damage Cytotoxic drugs Radiation BAX gene expression Increased BAX protein p53 Figure 1.9 Representation of apoptosis. Apoptosis is initiated via two main stimuli: (i) signalling through cell membrane receptors such as FAS or tumour necrosis factor (TNF) receptor; or (ii) release of cytochrome c from mitochondria. Membrane receptors signal apoptosis through an intracellular death domain leading to activation of caspases which digest DNA. Cytochrome c binds to the cytoplasmic protein Apaf-1 leading to activation of caspases. The intracellular ratio of pro-apoptotic, e.g. BAX, or anti- apoptotic, e.g. BCL-2, members of the BCL-2 family may influence mitochondrial cytochrome c release. Growth factors raise the level of BCL-2, inhibiting cytochrome c release, whereas DNA damage, by activating p53, raises the level of BAX, which enhances cytochrome c release.
10 / Chapter 1: Haemopoiesis proteins that are involved in mediating apoptosis following DNA damage, such as p53 and ATM, are also frequently mutated and therefore inactivated in haemopoietic malignancies. Necrosis is death of cells and adjacent cells due to ischemia, chemical trauma or hyperthermia. The cells swell and the plasma membrane loses integrity. There is usually an inflam- matory infiltrate in response to spillage of cell contents. Autophagy is the digestion of cell organelles by lysosomes. It may be involved in cell death, but in some situations also in maintaining cell survival by recycling nutrients. Now visit www.wiley.com/go/haematology9e to test yourself on this chapter. ■ Haemopoiesis (blood cell formation) arises from pluripotent stem cells in the bone marrow. Haemopoietic stem cells give rise to mixed and then single lineage progenitor and precursor cells which, after multiple cell divisions and differentiation, form red cells, granulocytes (neutrophils, eosinophils and basophils), monocytes, platelets, B and T lymphocytes and natural killer (NK) cells. ■ Haemopoietic tissue occupies about 50% of the marrow space in normal adult marrow. Haemopoiesis in adults is confined to the central skeleton, but in infants and young children haemopoietic tissue extends down the long bones of the arms and legs. ■ Stem cells reside in the bone marrow in osteoblastic or endothelial niches formed by stromal cells. They also circulate in the blood. ■ Growth factors attach to specific cell surface receptors and produce a cascade of phosphorylation events in the cell nucleus. ■ Transcription factors are molecules that bind to DNA and control the transcription of specific genes or gene families. They carry the message to those genes that are to be ‘switched on or off’ , to stimulate cell division, differentiation or functional activity or to suppress apoptosis. ■ Adhesion molecules are a large family of glycoproteins that mediate the attachment of marrow precursors and mature leucocytes and platelets to extracellular matrix, to endothelium and to each other. ■ Epigenetics refers to changes in DNA and chromatin that affect gene expression other than those that affect DNA sequence. Histone modification and DNA (cytosine) methylation are two important examples relevant to haemopoiesis and haematological malignancies. ■ Apoptosis is a physiological process of cell death resulting from activation of caspases. The intracellular ratio of pro-apoptotic proteins, e.g. BAX, to anti- apoptotic proteins, e.g. BCL-2, determines the cell susceptibility to apoptosis. SUMMARY
12 / Chapter 2: Erythropoiesis and general aspects of anaemia Table 2.1 The blood cells. Cell Diameter (μm) Life span in blood Number Function Red cells 6–8 120 d Male: 4.5–6.5 × 10 12 /L Female: 3.9–5.6 × 10 12 /L Oxygen and carbon dioxide transport Platelets 0.5–3.0 10 d 140–400 × 10 9 /L Haemostasis Phagocytes Neutrophils 12–15 6–10 h 1.8–7.5 × 10 9 /L Protection from bacteria, fungi Monocytes 12–20 20-40 h 0.2–0.8 × 10 9 /L Protection from bacteria, fungi Eosinophils 12–15 Days 0.04–0.44 × 10 9 /L Protection against parasites Basophils 12–15 Days 0.01–0.1 × 10 9 /L Lymphocytes B T 7–9 (resting) 12–20 (active) Weeks or years 1.5–3.5 × 10 9 /L B cells: immunoglobulin synthesis T cells: protection against viruses; immune functions Natural killer cells NK 10 (resting) 10–20 (active) Hours or days 0.1–0.4 Protection against virus- infected and neoplastic cells Blood cells All the circulating blood cells derive from pluripotential stem cells in the marrow. They divide into three main types. The most numerous are red cells (erythrocytes), specialized for the car- riage of oxygen from the lungs to the tissues and of carbon diox- ide in the reverse direction (Table 2.1). They have a 4-month life span, whereas the smallest cells, platelets involved in haemosta- sis, circulate for only 10 days. The white cells are made up of four main types of phagocyte: neutrophils, eosinophils, basophils and monocytes, which protect against bacterial and fungal infections (Chapter 8); and of lymphocytes, which include B cells, involved in antibody production, T cells (CD4 helper and CD8 suppressor), concerned with the immune response and in protection against viruses and foreign cells, and natural killer (NK) cells, a subset of CD8 T cells (Chapter 9). White cells have a wide range of life span (Table 2.1). The red cells and platelets are counted and their diameter and other parameters measured by an automated cell counter (Fig. 2.1). The counter also enumerates the different types of white cell by flow cytometry and detects abnormal cells. Erythropoiesis We each make approximately 10 12 new erythrocytes each day by the complex and finely regulated process of erythropoiesis. This progresses from the stem cell through progenitor cells, the erythroid and megakaryocyte colony-forming unit (CFU MkE ), burst-forming unit erythroid (BFU E ) and erythroid CFU (CFU-E; Fig. 1.2) to the first recognizable erythrocyte precur- sor in the bone marrow, the pronormoblast (Fig. 2.2). This process occurs in an erythroid niche in which about 30 eryth- roid cells at various stages of development surround a central macrophage. The pronormoblast is a large cell with dark blue cytoplasm, a central nucleus with nucleoli and slightly clumped chromatin (Fig. 2.2). It gives rise to a series of progressively smaller nor- moblasts by a number of cell divisions. These also contain
Chapter 2: Erythropoiesis and general aspects of anaemia / 13 Automated cell counter Neutrophils Computer screen Printer Whole blood in EDTA Bar code Bar code reader Specimen tubes in rack on a track Platelet count and size Red cell count and size Flow cytometry – white cell differentiatial Lysis of red cells Eosinophils Basophils Monocytes Lymphocytes Figure 2.1 Automated blood cell counter. Source: A.B. Mehta, A.V. Hoffbrand (2014) Haematology at a Glance, 4th edn. Reproduced with permission of John Wiley & Sons. (b) (a) (d) (c) Figure 2.2 Erythroblasts (normoblasts) at varying stages of development. The earlier cells are larger, with more basophilic cytoplasm and a more open nuclear chromatin pattern (a, b). The cytoplasm of the later cells is paler blue and more eosinophilic as a result of haemoglobin formation (c, d).
14 / Chapter 2: Erythropoiesis and general aspects of anaemia progressively more haemoglobin (which stains pink) in the cytoplasm; the cytoplasm also stains paler blue as it loses its RNA and protein synthetic apparatus, while nuclear chroma- tin becomes more condensed (Figs. 2.2 and 2.3). The nucleus is finally extruded from the late normoblast within the marrow and a reticulocyte results. This still contains some ribosomal RNA and so is still able to synthesize haemoglobin (Fig. 2.4). The reticulocyte is slightly larger than a mature red cell. It circulates in the peripheral blood for 1–2 days before maturing, when RNA is completely lost. A completely pink- staining mature erythrocyte results, which is a non-nucleated biconcave disc (Fig. 2.4). One pronormoblast usually gives rise to 16 mature red cells (Fig. 2.3). Normoblasts are not present in normal human peripheral blood (Fig. 2.4). They appear in the blood if erythropoiesis is occurring outside the marrow (extramedullary erythropoiesis) and also with some marrow diseases. Erythropoietin Erythropoiesis is regulated by the hormone erythropoietin, a heavily glycosylated polypeptide. Ninety percent of the hormone is produced in the peritubular interstitial cells of the kidney and 10% in the liver and elsewhere. There are no pre- formed stores. The stimulus to erythropoietin production is the oxygen (O 2 ) tension in the tissues of the kidney (Fig. 2.5a). Erythropoietin production increases with decreased O 2 deliv- ery to the kidney. This is caused most frequently by anaemia, but also occurs when haemoglobin for some metabolic or structural reason is unable to give up O 2 normally, when atmospheric O 2 is low or with defective cardiac or pulmonary function or damage to the renal circulation. Hypoxia induces stabilization of the hypoxia-inducible factor (HIF-1α) which then forms a dimer with HIFβ, the dimer stimulating erythropoietin production. The dimer also Pronormoblast Bone Marrow Blood Early Intermediate (polychromatic) Late (pyknotic) Reticulocytes Red cells 60–80% in cell cycle Post-mitotic non-dividing Figure 2.3 The amplification and maturation sequence in the development of mature red cells from the pronormoblast. Normoblast Nuclear DNA RNA in cytoplasm In marrow In blood Yes Yes Yes No No Yes Yes Yes No Reticulocyte Mature RBC No Yes Yes Figure 2.4 Comparison of the DNA and RNA content, and marrow and peripheral blood distribution, of the erythroblast (normoblast), reticulocyte and mature red blood cell (RBC).
Chapter 2: Erythropoiesis and general aspects of anaemia / 15 stimulates new vessel formation, glycolytic enzyme and trans- ferrin receptor synthesis and increased iron absorption by reducing hepcidin synthesis. Prolyl hydroxylase (PHD2) is a key oxygen sensor. It uses molecular oxygen to hydroxylate HIFα. Hydroxylation allows the von Hippel-Lindau (vHL) protein to break down HIFα by ubiquitination (Fig. 2.5b). Mutations in the genes vHL, PHD2 and HIF2α are rare causes of congenital polycythaemia (Chapter 15). Daprodustat, Stem cells Early BFU-E Late BFU-E Bone marrow Kidney CFU-E Reticulocyte O 2 delivery O 2 sensor (HIFα and β) Atmospheric O 2 O 2 -dissociation curve Cardiopulmonary function Haemoglobin concentration Renal circulation (Pro)normoblasts Erythropoietin Circulating red cells Peritubular interstitial cells of outer cortex Figure 2.5a The production of erythropoietin by the kidney in response to its oxygen (O 2 ) supply. Erythropoietin stimulates erythropoiesis and so increases O 2 delivery. BFU E , erythroid burst-forming unit; CFU-E, erythroid colony-forming unit. Prolyl hydroxylase oxygen sensor vHL OH Ub Ub Ub Ub Ub Ub HIF1-α EpoR Erythropoiesis TfR Supply of iron VEGF Angiogenesis Glycolytic enzymes Supply of energy Degradation via proteasome Oxygenation HIF1-α HIF1-β Hypoxia High O 2 Figure 2.5b The oxygen sensor: hypoxia stabilises hypoxia inducible factor (HIF)α which then forms a dimer with HIFβ, which stimulates erythropoietin production. PHD2 (prolyl hydroxylase), the oxygen sensor, uses molecular oxygen to hydroxylate HIF-1α. This hydroxylation allows von Hippel-Lindau (vHL) binding to HIFα and stimulates its breakdown by ubiquitination. Source: D.R. Higgs et al. In A.V. Hoffbrand et al. (eds) (2016) Postgraduate Haematology, 7th edn. Reproduced with permission of John Wiley & Sons.
16 / Chapter 2: Erythropoiesis and general aspects of anaemia roxadustat and vadadustat which inhibit PDH2 and raise endogenous erythropoietin production are in clinical trials for treating the anaemia of chronic renal failure and as a result of chemotherapy for cancer. Erythropoietin stimulates erythropoiesis by increas- ing the number of progenitor cells committed to erythropoiesis. The transcription factor GATA2 is involved in initiating erythroid differentiation from pluripotential stem cells. Subsequently the transcription factors GATA1 and FOG1 are activated by erythropoietin receptor stimulation and are important in enhancing expression of erythroid-specific genes, e.g. of globin, haem biosynthetic and red cell membrane pro- teins, and also enhancing expression of anti-apoptotic genes and of the transferrin receptor1 (CD71). Late BFU E and CFU-E, which have erythropoietin receptors, are stimulated to proliferate, differentiate and produce haemoglobin. The pro- portion of erythroid cells in the marrow increases and, in the chronic state, there is anatomical expansion of erythropoiesis into fatty marrow and sometimes into extramedullary sites. In infants, the marrow cavity may expand into cortical bone, resulting in bone deformities with frontal bossing and protru- sion of the maxilla (Chapter 7). Conversely, increased O 2 supply to the tissues (because of an increased red cell mass or because haemoglobin is able to release its O 2 more readily than normal) reduces the erythropoietin drive. Plasma erythropoietin levels can be valuable in clinical diagnosis. They are high in anaemia, unless this is due to renal failure or if a tumour-secreting erythropoietin is present, but low in severe renal disease or polycythaemia vera (Fig. 2.6). Indications for erythropoietin therapy Recombinant erythropoietin is needed for treating anaemia resulting from renal disease or from various other causes. It is given subcutaneously either three times weekly, once every 1–2 weeks or every 4 weeks, depending on the indication and on the preparation used (erythropoietin alpha or beta; darbepoetin alpha, a heavily glycosylated longer-acting form; or Micera, the longest-acting preparation). The main indication is end-stage renal disease (with or without dialysis). The patients often also need oral or intravenous iron. Other indications are listed in Table 2.2. The haemoglobin level and quality of life may be improved. A low serum erythropoietin level prior to treatment is valuable in predicting an effective response. Side effects include a rise in blood pressure, thrombosis and local injection site reac- tions. Erythropoietin has been associated with progression of some tumours which express EPO receptors and so with reduced survival. It is only indicated as an alternative to blood transfusion in cancer patients with symptomatic anaemia where the benefits outweigh the risks of tumour progression and of venous throm- bosis. Prolyl hydroxylase inhibitors (see above) are undergoing trials for treating anaemia in cancer patients. The marrow requires many other substances for effective erythropoiesis. These include metals iron and cobalt, vitamins (vitamin B 12 , folate, vitamin C, vitamin E, vitamin B 6 , thiamine and riboflavin) and hormones androgens and thyroxine. Deficiency in any of these may be associated with anaemia. 10 1 40 60 80 100 120 140 160 Haemoglobin in g/L 180 10 4 10 2 10 3 EPO (mIU/mL) 10 5 Renal failure: Normal Anaemias Nephric Anephric Figure 2.6 The relation between the concentration of erythropoietin (EPO) in plasma and haemoglobin concentration. Anaemias in this figure exclude conditions shown to be associated with impaired production of EPO. Source: Modified from M. Pippard et al. (1992) Br. J. Haematol. 82: 445. Reproduced with permission of John Wiley & Sons. Table 2.2 Clinical indications (in selected subjects) for erythropoietin. Anaemia of chronic renal disease Myelodysplastic syndrome Anaemia associated with malignancy and chemotherapy Anaemia of chronic diseases, e.g. rheumatoid arthritis Anaemia of prematurity Perioperative uses
Chapter 2: Erythropoiesis and general aspects of anaemia / 17 Haemoglobin Haemoglobin synthesis Each molecule of normal adult haemoglobin A (Hb A, the dominant haemoglobin in blood after the age of 3–6 months) consists of four polypeptide chains, α 2 β 2 , each with its own haem group. Normal adult blood also contains small quantities of two other haemoglobins: Hb F and Hb A 2 . These also con- tain α chains, but with γ and δ chains, respectively, instead of β (Table 2.3). The synthesis of the various globin chains in the foetus and adult is discussed in Chapter 7. Haem synthesis occurs largely in mitochondria by a series of biochemical reactions, commencing with the condensation of glycine and succinyl coenzyme A under the action of the key rate-limiting enzyme δ-aminolaevulinic acid synthase (ALAS) (Fig. 2.7). Pyridoxal phosphate (vitamin B 6 ) is a coenzyme for this reaction. The main sources of succinyl CoA are glutamine and glucose, which are converted to alpha-ketoglutarate, a succinate precursor inside the erythroid cells. Ultimately, pro- toporphyrin combines with iron in the ferrous (Fe 2+ ) state to form haem (Fig. 2.8). A tetramer of four globin chains, each with its own haem group in a ‘pocket’, is then formed to make up a haemoglobin molecule (Fig. 2.9). An enzyme eIF2alpha kinase, also known as haem-regulated inhibitor (HRI), senses intracellular haem concentration. If this is low as in iron defi- ciency, HRI phosphorylates its substrate eIF2alpha which then reduces globin synthesis by inhibiting its mRNA translation. An adjacent gene Nprl3 shares some enhancers with the α-globin gene and so the control of expression of the two genes is coupled. Nprl3 provides negative regulation of mTORC1, a critical controller of cellular metabolism. Nprl3 is essential for optimal erythropoiesis and for responding to fluctuating nutri- ent (including iron) and growth factor concentrations. Mitochondrion Ribosomes Amino acids Haemoglobin Transferrin Transferrin cycle Ferritin Fe Porphobilinogen Uroporphyrinogen Coproporphyrinogen Proto- porphyrin Haem (x4) Glycine + B6 + Succinyl CoA δALA α 2 β 2 globin α and β chains Fe Figure 2.7 Haemoglobin synthesis in the developing red cell. The mitochondria are the main sites of protoporphyrin synthesis, iron (Fe) is supplied from circulating transferrin and globin chains are synthesized on ribosomes. δ-ALA, δ-aminolaevulinic acid; CoA, coenzyme A. Table 2.3 Normal haemoglobins in adult blood. Hb A Hb F Hb A 2 Structure α 2 β 2 α 2 γ 2 α 2 δ 2 Normal (%) 96–98 0.5–0.8 1.5–3.2 Haem O 2 O 2 O 2 O 2 2,3-DPG α 1 β 1 β 2 α 2 Oxyhaemoglobin Deoxyhaemoglobin α 1 β 1 β 2 α 2 Figure 2.9 The oxygenated and deoxygenated haemoglobin molecule. α, β, globin chains of normal adult haemoglobin (Hb A); 2,3-DPG, 2,3-diphosphoglycerate. N Fe N CH 3 CH CH 2 CH CH 3 H 3 C H 3 C CH 2 βCH HCδ CH 2 N N C H γ C H α CH 2 COOH CH 2 CH 2 COOH Globin Figure 2.8 The structure of haem.
18 / Chapter 2: Erythropoiesis and general aspects of anaemia Haemoglobin function The red cells in systemic arterial blood carry O 2 from the lungs to the tissues and return in venous blood with CO 2 to the lungs. As the haemoglobin molecule loads and unloads O 2 , the individual globin chains move on each other (Fig. 2.9). The α 1 β 1 and α 2 β 2 contacts stabilize the molecule. When O 2 is unloaded the β chains are pulled apart, permitting entry of the metabolite 2,3-diphosphoglycerate (2,3-DPG), resulting in a lower affinity of the molecule for O 2 . This movement is responsible for the sigmoid form of the haemoglobin O 2 dissociation curve (Fig. 2.10). The P 50 (the partial pressure of O 2 at which haemoglobin is half satu- rated with O 2 ) of normal blood is 26.6 mmHg. With increased affinity for O 2 , the curve shifts to the left (the P 50 falls), while with decreased affinity for O 2 , the curve shifts to the right (the P 50 rises). Normally, in vivo, O 2 exchange operates between 95% sat- uration (arterial blood) with a mean arterial O 2 tension of 95 mmHg and 70% saturation (venous blood) with a mean venous O 2 tension of 40 mmHg (Fig. 2.10). The normal position of the curve depends on the concen- tration of 2,3-DPG, H + ions and CO 2 in the red cell and on the structure of the haemoglobin molecule. High concentrations of 2,3-DPG, H + or of CO 2 , and the presence of sickle haemo- globin (Hb S), shift the curve to the right (oxygen is given up more easily), whereas foetal haemoglobin (Hb F) – which is unable to bind 2,3-DPG – and certain rare abnormal haemoglobins associated with polycythaemia shift the curve to the left because they give up O 2 less readily than normal. Methaemoglobinaemia This is a clinical state in which circulating haemoglobin is present with iron in the oxidized Fe 3+ instead of the usual Fe 2+ state. It may arise because of a hereditary deficiency of the enzyme methaemoglobin reductase or inheritance of a struc- turally abnormal haemoglobin (Hb M). Hb Ms contain an amino acid substitution affecting the haem pocket of the globin chain. Toxic methaemoglobinaemia and/or sulphaemo- globinaemia occurs when a drug or other toxic substance oxidizes haemoglobin. In all these states, the patient is likely to show cyanosis. The red cell In order to carry haemoglobin into close contact with the tissues and for successful gaseous exchange, the red cell, 8 μm in diameter, must pass repeatedly through the microcircula- tion, whose minimum diameter is 3.5 μm. It must maintain haemoglobin in a reduced (ferrous) state and maintain osmotic equilibrium despite the high concentration of pro- tein (haemoglobin) in the cell. A single journey round the body takes 20 seconds and its total journey throughout its 120-day life span has been estimated to be 480 km (300 miles). To fulfil these functions, the cell is a flexible biconcave disc with an ability to generate energy as adeno- sine triphosphate (ATP) by the anaerobic glycolytic (Embden–Meyerhof ) pathway (Fig. 2.11) and to generate reducing power both as nicotinamide adenine dinucleotide (NADH) by this pathway and as reduced nicotinamide ade- nine dinucleotide phosphate (NADPH) by the hexose monophosphate shunt (Fig. 2.11, Fig. 6.6). Red cell metabolism Embden–Meyerhof pathway In this series of biochemical reactions, glucose that enters the red cell from plasma by facilitated transfer is metabolized to lactate (Fig. 2.11). For each molecule of glucose used, two mol- ecules of ATP and thus two high-energy phosphate bonds are generated. This ATP provides energy for maintenance of red cell volume, shape and flexibility. The Embden–Meyerhof pathway generates NADH, which is needed by the enzyme methaemoglobin reductase to reduce functionally dead methaemoglobin containing ferric iron, produced by oxidation of approximately 3% of haemoglobin each day, to functionally active haemoglobin containing ferrous ions. The Rapoport–Luebering shunt, or side-arm, of this pathway (Fig. 2.11) generates 2,3-DPG, important in the regulation of haemoglobin’s oxygen affinity (Fig. 2.10). Mean venous O 2 tension P 50 Arterial O 2 tension 0 0 25 50 75 PO 2 100 75 25 50 % saturation haemoglobin 100 2,3-DPG H + HbF 2,3-DPG H + CO 2 HbS Figure 2.10 The haemoglobin oxygen (O2) dissociation curve. 2,3- DPG, 2,3-diphosphoglycerate.
Chapter 2: Erythropoiesis and general aspects of anaemia / 19 Hexose monophosphate (pentose phosphate) shunt Approximately 10% of glycolysis occurs by this oxidative pathway in which glucose-6-phosphate is converted to 6-phosphogluconate and so to a pentose-5-phosphate (Fig. 2.11, Fig. 6.6). NADPH is generated and is linked with glutathione, which maintains sulphydril (SH) groups intact in the cell, including those in haemoglobin and in the red cell membrane. In one of the most common inherited abnormalities of red cells, glucose-6-phosphate dehydrogenase (G6PD) deficiency, the red cells are extremely susceptible to oxidant stress (Chapter 6). Red cell membrane The red cell membrane comprises a lipid bilayer, membrane integral proteins and a membrane skeleton (Fig. 2.12). Approximately 50% of the membrane is protein, 20% phospholipids, 20% cholesterol molecules and up to 10% is carbohydrate. Carbohydrates occur only on the external sur- face, while proteins are either integral, penetrating the lipid bilayer, or form a skeleton on the inner surface of the mem- brane. Several red cell proteins have been numbered according to their mobility on polyacrylamide gel electrophoresis (PAGE), e.g. band 3, proteins 4.1, 4.2 (Fig. 2.12). GSH G6PD Glucose-6-phosphate GLUCOSE Embden–Meyerhof glycolytic pathway Rapoport– Luebering shunt Glyceraldehyde-3-phosphate 1,3-Diphosphoglycerate 2,3-Diphosphoglycerate (2,3-DPG) Effect on oxygen dissociation curve (Fig 2.10) 3-Phosphoglycerate Phosphoenolpyruvate Pyruvate Lactate PK 6-Phosphoglycerate Pentose phosphate pathway GS H 2 O Toxic oxygen species e.g. superoxide NADP NADPH Hb (Fe 2+ ) Met Hb (Fe 3+ ) NAD + NADH ADP ATP ADP ATP Figure 2.11 The anaerobic Embden–Meyerhof pathway generates energy as ATP and reducing power as NADH. The pentose-phosphate shunt pathway generates additional reducing power as NADPH. Further down the main pathway the Rapoport–Luebering shunt generates 2,3 DPG which effects oxygen binding and release by haemoglobin (Fig 2.10). GS, glutathione; GSH, reduced glutathione; G6PD, glucose-phosphate dehydrogenase; PK, pyruvate kinase; NAD, NADP, ADP, ATP see text.
20 / Chapter 2: Erythropoiesis and general aspects of anaemia The membrane skeleton is formed by structural proteins that include α and β spectrin, ankyrin, protein 4.1 and actin. These proteins form a horizontal lattice important in maintaining the biconcave shape. Spectrin is the most abundant and consists of two chains, α and β, wound around each other to form heterodi- mers, which then self-associate head to head to form tetramers. These tetramers are linked at the tail end to actin and attached there to protein band 4.1. At the head end, the β spectrin chains attach to ankyrin, which connects them to band 3, the trans- membrane protein that acts as an anion channel (‘vertical con- nection’; Fig. 2.12). Protein 4.2 enhances this interaction. Defects of the membrane proteins explain some of the abnormalities of shape of the red cell membrane, e.g. heredi- tary spherocytosis and elliptocytosis (Chapter 6), while altera- tions in lipid composition because of congenital or acquired abnormalities in plasma cholesterol or phospholipid may be associated with other membrane abnormalities (Fig. 2.16). Anaemia Anaemia is defined as a reduction in the haemoglobin concentration of the blood below normal for age and sex (Table2.4). Although normal values can vary between labora- tories, typical values would be less than 135 g/L in adult males and less than 115 g/L in adult females (Fig. 2.13). The World Health Organisation (WHO) defines anaemia as a haemoglobin level < 130 g/L for adult males, < 120 g/L for adult non-pregnant females and <110 g/L from the age of 6–59 months. Newborn infants have a high haemoglobin level; 140 g/L is taken as the lower limit at birth (Fig. 2.13). Anaemia in pregnancy and neonates is discussed in Chapter 34. Alterations in total circulating plasma volume as well as in total circulating haemoglobin mass determine the haemo- globin concentration. Reduction in plasma volume as in dehydration may mask anaemia or even cause apparent (pseudo) polycythaemia (Chapter 15). Conversely, an increase in plasma volume as with splenomegaly or pregnancy may cause anaemia even with a normal total circulating red cell and haemoglobin mass. After acute major blood loss, anaemia is not immediately apparent because the total blood volume is reduced. It takes up to a day for the plasma volume to be replaced and so for the degree of anaemia to become apparent. Regeneration of red cells and haemoglobin mass takes substantially longer. The ini- tial clinical features of major blood loss are therefore a result of reduction in blood volume rather than of anaemia. Global incidence On the basis of the WHO definitions, anaemia was estimated in 2010 to occur in about 33% of the global population. Prevalence was greater in females than males at all ages and most frequent in children less than 5 years old. Anaemia was most frequent in Horizontal interaction Vertical interaction 4.2 Glycophorin B Membrane phospholipid Band 3 protein Glycophorin A Glycophorin C Cholesterol Cytoskeleton Actin α Spectrin Ankyrin β Spectrin Tropomyosin 4.1 4.1 Figure 2.12 The structure of the red cell membrane. Some of the penetrating and integral proteins carry carbohydrate antigens; other antigens are attached directly to the lipid layer.
Chapter 2: Erythropoiesis and general aspects of anaemia / 21 South Asia, and in Central, West and East Sub-Saharan Africa. The main causes are iron deficiency (caused by life-long poor diet com- bined with menstruation and/or repeated pregnancies, hookworm, schistosomiasis), the anaemia of inflammation (Chapter 3), sickle cell diseases, thalassaemia (Chapter 7), malaria (Chapter 32). Clinical features of anaemia The major adaptations to anaemia are in the cardiovascular system with increased cardiac stroke volume and tachycar- dia and in the haemoglobin O 2 dissociation curve. In some patients with quite severe anaemia, there may be no symptoms or signs, whereas others with mild anaemia may be severely incapacitated. The presence or absence of clinical features depends on: 1 Speed of onset Rapidly progressive anaemia causes more symptoms than anaemia of slow onset. This is because there is less time for adaptation in the cardiovascular system and in the O 2 dissociation curve of haemoglobin. 2 Severity Mild anaemia often produces no symptoms or signs, but these are usually present when the haemoglobin is less than 90 g/L. Even severe anaemia (haemoglobin con- centration as low as 60 g/L) may produce remarkably few symptoms, when there is very gradual onset in young sub- jects who are otherwise healthy. 3 Age The elderly tolerate anaemia less well than the young because normal cardiovascular compensation is impaired. 4 Haemoglobin O 2 dissociation curve Anaemia, in gen- eral, is associated with a rise in 2,3-DPG in the red cells and a shift in the O 2 dissociation curve to the right, so that oxygen is given up more readily to tissues. This adaptation, which takes days to occur, is particularly marked in some anaemias that either raise 2,3-DPG directly, e.g. pyruvate kinase deficiency (Chapter 6), or that are associated with a low-affinity haemoglobin, e.g. Hb S (Fig. 2.10). Symptoms If the patient does have symptoms, these are usually shortness of breath, particularly on exertion, weakness, lethargy, palpita- tion and headaches. In older subjects, symptoms of cardiac fail- ure, angina pectoris, intermittent claudication or confusion may be present. Visual disturbances because of retinal haemor- rhages may complicate very severe anaemia, particularly of rapid onset (Fig. 2.14). Table 2.4 Normal values for blood cells and haematinics. Males Females Haemoglobin (g/L) 135.0–175.0 115.0–155.0 Red cells (erythrocytes) (×10 12 /L) 4.5–6.5 3.9–5.6 PCV (haematocrit) (%) 40–52 36–48 Mean cell volume (MCV) (fL) 80–95 Mean cell haemoglobin (MCH) (pg) 27–34 Reticulocyte count (×10 9 /L) 50–150 White cells (leucocytes) Total (×10 9 /L) 4.0–11.0 Neutrophils (×10 9 /L) 1.8–7.5 (Caucasians 1.5–7.5 (Africa and Middle- East) Lymphocytes (×10 9 /L) 1.5–3.5 Monocytes (×10 9 /L) 0.2–0.8 Eosinophils (×10 9 /L) 0.04–0.44 Basophils (×10 9 /L) 0.01–0.1 Platelets (×10 9 /L) 150–400 Serum iron (μmol/L) 10–30 Total iron-binding capacity (μmol/L) 40–75 (2.0–4.0 g/L as transferrin) Serum ferritin* (μg/L) 40–340 14–150 Serum vitamin B 12 * (ng/L) 160–925 (20–680 pmol/L) Serum folate* (μg/L) 3.0–15.0 (4–30 nmol/L) Red cell folate** (μg/L) 160–640 (360–1460 nmol/L) PCV, packed cell volume. * Normal ranges differ between laboratories. ** Normal ranges differ between different laboratories. 0 Years Months Age: 120 110 Haemoglobin (g/L) 3 2 1 1 5 10 Men Neonates Infants Children Women 20 30 40 50 60 70 130 140 100 90 Figure 2.13 The lower limit of blood haemoglobin concentration in healthy men, women and children of various ages.
22 / Chapter 2: Erythropoiesis and general aspects of anaemia Signs These may be divided into general and specific. General signs include pallor of mucous membranes or nail beds, which occurs if the haemoglobin level is less than 90 g/L (Fig. 2.15). Conversely, skin colour is not a reliable sign. A hyperdynamic circulation may be present with tachycardia, a bounding pulse, cardiomegaly and a systolic flow murmur. Particularly in the elderly, features of congestive heart failure may be present. Specific signs are associated with particular types of anae- mia, e.g. koilonychia (spoon nails) with iron deficiency, jaun- dice with haemolytic or megaloblastic anaemias, leg ulcers with sickle cell and other haemolytic anaemias, or bone deformities with thalassaemia major. The association of features of anaemia with excess infec- tions or spontaneous bruising suggests that neutropenia or thrombocytopenia may be present, possibly as a result of bone marrow failure. Classification of anaemia Red cell indices The most useful classification is based on the red cell indices, especially MCV. This divides the anaemia into microcytic, nor- mocytic and macrocytic (Table 2.5). As well as suggesting the nature of the primary defect, this classification may also indicate an underlying abnormality before overt anaemia has developed. In two common physiological situations, the mean corpus- cular volume (MCV) may be outside the normal adult range. In the newborn for a few weeks, the MCV is high, but in infancy it is low, e.g. 70 fL at 1 year of age and rises slowly throughout childhood to the normal adult range. In normal pregnancy there is a slight rise in MCV, even in the absence of other causes of macrocytosis, e.g. folate deficiency. Other laboratory findings Although the red cell indices will indicate the type of anaemia, further useful information can be obtained from the initial blood sample. Leucocyte and platelet counts Measurement of these helps to distinguish ‘pure’ anaemia from ‘pancytopenia’ (subnormal levels of red cells, neutrophils and platelets), which suggests a more general marrow defect or destruction of cells, e.g. hypersplenism. In anaemias caused by haemolysis or haemorrhage, the neutrophil and platelet counts are often raised; in infections and leukaemias, the leucocyte count is also often raised, and there may be abnormal leuco- cytes or neutrophil precursors present. Reticulocyte count The normal percentage is 0.5–2.5%, and the absolute count 50–150 × 10 9 /L (Table 2.4). This should rise in anaemia because of erythropoietin increase and be higher the more Figure 2.14 Retinal haemorrhages in a patient with severe anaemia (haemoglobin 25 g/L) caused by severe haemorrhage. (b) (a) Figure 2.15 Pallor of the conjunctival mucosa (a) and of the nail bed (b) in two patients with severe anaemia (haemoglobin 60 g/L).
Chapter 2: Erythropoiesis and general aspects of anaemia / 23 severe the anaemia. This is particularly so when there has been time for erythroid hyperplasia to develop in the marrow as in chronic haemolysis. After an acute major haemorrhage, there is an erythropoietin response in 6 hours, and the reticulocyte count rises within 2–3 days, reaches a maximum in 6–10 days and remains raised until the haemoglobin returns to the nor- mal level. If the reticulocyte count is not raised in an anaemic patient, this suggests impaired marrow function, lack of eryth- ropoietin (renal disease) or lack of erythropoietin stimulus (Table 2.6). Blood film It is important to examine the blood film in all cases of anae- mia. Abnormal red cell morphology (Fig. 2.16) or red cell inclusions (Fig. 2.17) may suggest a particular diagnosis. During the blood film examination, white cell abnormalities are sought, platelet number and morphology assessed and the presence of abnormal cells, e.g. normoblasts, granulocyte pre- cursors or blast cells, is noted. Bone marrow examination This is needed when the cause of anaemia or other abnor- mality of the blood cells cannot be diagnosed from the blood count, film and other blood tests alone. It may be performed by aspiration or trephine biopsy (Fig.2.18). For bone marrow aspiration, a needle is inserted into the marrow cavity and a liquid sample of marrow is sucked into a syringe. This sample is then spread on a slide for microscopy and stained by the usual Romanowsky technique. The detail of the developing cells can be examined, e.g. normoblastic or megalo- blastic and the proportion of the different cell lines assessed (myeloid: erythroid ratio), the proportion of granulocyte pre- cursors to red cell precursors in the bone marrow, normally 2.5 : 1 to 12 : 1. The presence of cells foreign to the marrow, e.g. secondary carcinoma, can also be observed. The cellularity of the marrow can be viewed provided fragments are obtained. An iron stain is performed routinely so that the amount of iron in reticuloendothelial stores (macrophages) and as fine Table 2.5 Classification of anaemia. Microcytic, hypochromic Normocytic, normochromic Macrocytic MCV <80 fL MCV 80–95 fL MCV >95 fL MCH <27 pg MCH ≥27 pg Megaloblastic: vitamin B 12 or folate deficiency. Non-megaloblastic: alcohol, liver disease, myelodysplasia, aplastic anaemia, etc. (Table 5.10) Iron deficiency Many haemolytic anaemias Thalassaemia Anaemia of chronic disease (some cases) Lead poisoning Sideroblastic anaemia (some cases) Anaemia of chronic disease (some cases) After acute blood loss Renal disease Mixed deficiencies Bone marrow failure e.g. post-chemotherapy, infiltration by carcinoma, etc. MCH, mean corpuscular haemoglobin; MCV, mean corpuscular volume. Table 2.6 Factors impairing the normal reticulocyte response to anaemia. Marrow diseases, e.g. hypoplasia, infiltration by carcinoma, lymphoma, myeloma, acute leukaemia, tuberculosis Deficiency of iron, vitamin B 12 or folate Lack of erythropoietin, e.g. renal disease Reduced tissue O 2 consumption, e.g. myxoedema, protein deficiency Ineffective erythropoiesis, e.g. thalassaemia major, megaloblastic anaemia, myelodysplasia, myelofibrosis Chronic inflammatory or malignant disease
24 / Chapter 2: Erythropoiesis and general aspects of anaemia granules (‘siderotic’ granules) in the developing erythroblasts can be assessed (Fig. 3.10). An aspirate sample may also be used for a number of other specialized investigations (Table 2.7). A trephine biopsy provides a solid core of bone including marrow and is examined as a histological specimen after fixa- tion in formalin, decalcification and sectioning. Usually immu- nohistology is performed, depending on the diagnosis suspected (Chapter 11). A trephine biopsy specimen is less valuable than aspirate when individual cell detail is to be examined, but pro- vides a panoramic view of the marrow, from which overall mar- row architecture, cellularity and presence of fibrosis or abnormal infiltrates can, with immunohistology if needed, be reliably determined. Ineffective erythropoiesis Erythropoiesis is not entirely efficient since approximately 10–15% of developing erythroblasts die within the marrow without producing mature cells. This is termed ineffective erythropoiesis and it is substantially increased in a number of chronic anaemias (Fig. 2.19). The serum unconjugated bili- rubin (derived from breaking down haemoglobin) and lac- tate dehydrogenase (LDH, derived from breaking down cells) are usually raised when ineffective erythropoiesis is marked. The reticulocyte count is low in relation to the degree of anaemia and to the proportion of erythroblasts in the marrow. Basket cell Red cell abnormality Causes Red cell abnormality Causes Macrocyte Liver disease, alcoholism. Oval in megaloblastic anaemia Target cell Iron deficiency, liver disease, haemoglobinopathies, post-splenectomy Stomatocyte Liver disease, alcoholism Pencil cell Iron deficiency Echinocyte Liver disease, post-splenectomy. storage artefact Acanthocyte Liver disease, abetalipo- proteinaemia, renal failure Sickle cell Sickle cell anaemia Microcyte Iron deficiency, haemoglobinopathy Microspherocyte Hereditary spherocytosis, autoimmune haemolytic anaemia, septicaemia Fragments DIC, microangiopathy, HUS, TTP, burns, cardiac valves Elliptocyte Hereditary elliptocytosis Tear drop poikilocyte Myelofibrosis, extramedullary haemopoiesis Oxidant damage– e.g. G6PD deficiency, unstable haemoglobin Normal Figure 2.16 Some of the more frequent variations in size (anisocytosis) and shape (poikilocytosis) that may be found in different anaemias. DIC, disseminated intravascular coagulopathy; G6PD, glucose-6-phosphate dehydrogenase; HUS, haemolytic-uraemic syndrome; TTP, thrombotic thrombocytopenic purpura.
Chapter 2: Erythropoiesis and general aspects of anaemia / 25 Assessment of erythropoiesis Total erythropoiesis and the amount of erythropoiesis that is effec- tive in producing circulating red cells can be assessed by examining the bone marrow, haemoglobin level and reticulocyte count. Total erythropoiesis is assessed from the marrow cellularity and the myeloid: erythroid ratio. This ratio falls and may be reversed when total erythropoiesis is selectively increased. Effective erythropoiesis is assessed by the reticulocyte count. This is raised in proportion to the degree of anaemia when erythropoiesis is effective, but is low when there is inef- fective erythropoiesis or an abnormality preventing normal marrow response (Table 2.6). Normoblast (nucleated RBC) Reticulocyte (RNA) Heinz bodies Howell-Jolly body Siderotic granules (Pappenheimer bodies) Basophilic stippling Malarial parasite Figure 2.17 Red blood cell (RBC) inclusions which may be seen in the peripheral blood film in various conditions. The reticulocyte RNA and Heinz bodies are only demonstrated by supravital staining, e.g., with new methylene blue. Heinz bodies are oxidized denatured haemoglobin. Siderotic granules (Pappenheimer bodies) contain iron. They are purple on conventional staining, but blue with Perls’ stain. The Howell–Jolly body is a DNA remnant. Basophilic stippling is denatured RNA. (b) (a) Figure 2.18 (a) The bone marrow aspiration needle and a smear made from a bone marrow aspirate. (b) The bone marrow trephine biopsy needle and a normal trephine biopsy section. Table 2.7 Indications for bone marrow aspiration and trephine biopsy. Aspiration Trephine biopsy Site Posterior iliac crest (sternum if obese; tibia in infants) Posterior iliac crest Stains Romanowsky; Perls’ reaction (for iron) Haematoxylin and eosin; reticulin (silver stain) Result available 1–2 h 1–7 d (according to decalcification method) Indications Investigation of unexplained anaemia, neutropenia, thrombocytopenia, suspicion of leukaemia, myeloproliferative disorders, myelodysplasia, aplastic anaemia, lymphoma, myeloma, amyloid, secondary carcinoma, cases of splenomegaly or pyrexia of undetermined cause Special tests Flow cytometry, cytogenetics, FISH and molecular tests including DNA or RNA analysis for gene abnormalities. Consider microbiological culture, cytochemical markers and progenitor cell culture Immunohistology FISH, fluorescence in situ hybridization.
26 / Chapter 2: Erythropoiesis and general aspects of anaemia Days Normal Erythroid hypoplasia, e.g. aplastic anaemia Peripheral blood Marrow 120 4 Erythroid hyperplasia, e.g. haemolytic anaemia Ineffective erythropoiesis, e.g. megalo- blastic anaemia Figure 2.19 The relative proportions of marrow erythroblastic activity, circulating red cell mass and red cell lifespan in normal subjects and in three types of anaemia. ■ Erythropoiesis (red cell production) is regulated by erythropoietin, which is secreted by the kidney in response to hypoxia. Erythropoiesis occurs from mixed progenitor cells through a series of nucleated red cell precursors (normoblasts) to a reticulocyte stage, containing RNA but not DNA. ■ Various short- or long-acting preparations of erythropoietin are used clinically to treat anaemia in renal failure and other diseases. ■ Haemoglobin is the main protein in red cells. It consists of four polypeptide (globin) chains, in adults dominantly 2α and 2β, each containing an iron atom bound to protoporphyrin to form haem. ■ The red cell has two biochemical pathways for metabolizing glucose, the Embden–Meyerhof pathway, which generates ATP, needed for maintenance of red cell shape and flexibility, and NADH, which prevents oxidation of haemoglobin; and the hexose monophosphate pathway, which generates NADPH, important for maintaining glutathione, which protects haemoglobin and proteins in the red cell membrane from oxidation. ■ The Luebering–Rapoport shunt, or side arm, of the Embden–Meyerhof pathway generates 2,3-DPG, important in the regulation of haemoglobin’s oxygen affinity ■ The red cell membrane consists of a lipid bilayer, proteins which form a membrane skeleton and are integral to the membrane, and carbohydrate surface antigens. ■ Anaemia is defined as a haemoglobin level in blood below the normal level for age and sex. It is classified according to the size of the red cells (MCV) into macrocytic, normocytic and microcytic. The reticulocyte count, morphology of the red cells and changes in the white cell and/or platelet count also help in the diagnosis of the cause of anaemia. ■ The general clinical features of anaemia include fatigue, headaches, shortness of breath on exertion, pallor of mucous membranes and tachycardia. ■ Other features relate to particular types of anaemia, e.g. jaundice, glossitis, leg ulcers, bone changes. ■ Bone marrow examination by aspiration or trephine biopsy may be important in the investigation of anaemia as well as of many other haematological diseases. Special tests, e.g. immunology by flow cytometry or immunohistology, cytogenetics or molecular genetics, can be performed on the cells or core of bone obtained. SUMMARY Now visit www.wiley.com/go/haematology9e to test yourself on this chapter.
28 / Chapter 3: Hypochromic anaemias Iron is one of the most common elements in the Earth’s crust, yet iron deficiency is the most common cause of anae- mia, affecting about 500 million people worldwide. It is particularly frequent in low-income populations, such as in sub-Saharan Africa or South Asia, where the diet is frequently of poor quality and parasites, e.g. hookworm or schistosomia- sis, which cause iron loss due to haemorrhage, may be present. Moreover, the body has limited ability to absorb iron. Iron deficiency is the major cause of a microcytic, hypochromic anaemia, in which the mean corpuscular volume (MCV) and mean corpuscular haemoglobin (MCH) are both reduced and the blood film shows small (microcytic) and pale (hypochromic) red cells. This appearance is caused by a defect in haemoglobin synthesis. The major differential diag- nosis of a microcytic, hypochromic anaemia is between iron deficiency, the anaemia of chronic disease, which are both dealt with in this chapter, and thalassaemia, which is considered in Chapter 7 (Fig. 3.1). Nutritional and metabolic aspects of iron Body iron distribution and transport The transport and storage of iron are largely mediated by three proteins: transferrin, transferrin receptor 1 (TFR1) and ferritin. Each transferrin molecule can contain up to two atoms of iron in the ferric form. Transferrin delivers iron to tissues that have transferrin receptors (TFR1s), especially erythroblasts in the bone marrow, which incorporate the iron into haemoglo- bin (Figs. 2.7 and 3.2). The transferrin is then reutilized. At the end of their life, red cells are broken down in the macrophages of the reticuloendothelial system and the iron released from haemoglobin enters the plasma and provides most of the iron attached to transferrin. Only a small proportion of plasma transferrin iron comes from dietary iron, absorbed each day through the duodenum. Iron in excess of that needed for hae- moglobin synthesis is also released from erythroblasts and erythrocytes to plasma transferrin. There is a diurnal variation in serum iron, highest in the mornings or noon and then fall- ing to lowest levels in the evening. Some iron is stored in the macrophages as ferritin and hae- mosiderin, the amount varying widely according to overall body iron status. Ferritin is a water-soluble protein–iron com- plex. It is made up of an outer protein shell, apoferritin, con- sisting of 22 subunits and an iron–phosphate–hydroxide core. It contains up to 20% of its weight as iron and is not visible by light microscopy. Specialized intracellular carrier proteins transfer iron to ferritin and from ferritin to phagolysosomes for its degradation and release of its iron. Haemosiderin is an insoluble protein–iron complex of varying composition containing approximately 37% iron by weight. It is derived from partial lysosomal digestion of fer- ritin molecules and is visible in macrophages and other cells by light microscopy after staining by Perls’ (Prussian blue) reaction (Fig. 3.10). Iron in ferritin and haemosiderin is in the ferric form. It is mobilized after reduction to the ferrous form. A copper-containing enzyme caeruloplasmin catalyses oxidation of the iron to the ferric form for binding to plasma transferrin. Most body iron is in haemoglobin. Iron is also present in muscles as myoglobin and in most body cells in iron-containing enzymes, e.g. cytochromes or catalase (Table 3.1). This tissue iron (a) Iron deficiency (b) Chronic inflammation or malignancy Sideroblastic anaemia Thalassaemia (α or β) Haemoglobin Haem + Globin Protoporphyrin Iron Figure 3.1 The causes of a hypochromic microcytic anaemia. These include lack of iron (iron deficiency) or of iron release from macrophages to serum (anaemia of chronic disease), failure of protoporphyrin synthesis (sideroblastic anaemia) or of globin synthesis (α- or β-thalassaemia). Lead also inhibits haem and globin synthesis. Table 3.1 The distribution of body iron. Amount of iron in average adult Male (g) Female (g) Percentage of total Haemoglobin 2.4 1.7 65 Ferritin and haemosiderin 1.0 (0.3–1.5) 0.3 (0–1.0) 30 Myoglobin 0.15 0.12 3.5 Haem enzymes, e.g., cytochromes, catalase, peroxidases, flavoproteins 0.02 0.015 0.5 Transferrin-bound iron 0.004 0.003 0.1
Chapter 3: Hypochromic anaemias / 29 is less likely to become depleted than haemosiderin, ferritin and haemoglobin in states of iron deficiency, but some reduc- tion of tissue haem-containing enzymes may also occur. Regulation of ferritin and transferrin receptor 1 synthesis The levels of ferritin, TFR1, δ-aminolaevulinic acid synthase (ALAS) and also of the divalent metal transporter 1 (DMT1), important in iron absorption, are linked to iron status, so that iron overload causes a rise in tissue ferritin and a fall in TFR1 and DMT1, whereas in iron deficiency ferritin and ALAS are low and TFR1 and DMT1 increased. This linkage arises through the binding of an iron regulatory protein (IRP) to iron response elements (IREs) on the ferritin, TFR1, ALAS and DMT1 mRNA molecules. Iron deficiency increases the ability of IRP to bind to the IREs, whereas iron overload reduces the binding. The site of IRP binding to the IREs determines whether the amount of the individual mRNAs and so protein produced is increased or decreased (Fig. 3.3). Upstream binding reduces translation, whereas downstream binding stabilizes the mRNA, increasing translation and so protein synthesis. When plasma iron is raised and transferrin is saturated, the amount of iron transferred to parenchymal cells e.g. those of the liver, endocrine organs and heart is increased and this is the basis of the pathological changes associated with iron loading condi- tions. There may also be free iron in plasma (non-transferrin bound iron) which is toxic to different organs (Chapter 4). Duodenum Daily absorption ~ 1 mg Daily loss ~ 1 mg Urine, faeces, nails, hair, skin Transferrin Plasma (4 mg) Bone marrow normoblasts (150 mg) Macrophages (0.5–1.5 g) Ineffective erythropoiesis Menstrual loss (haemorrhage) Circulating haemoglobin (1.7–2.4 g) Liver, other parenchymal cells and tissues, especially muscle myoglobin (600–650 mg) Figure 3.2 Daily iron cycle. Most of the iron in the body is contained in circulating haemoglobin and is reutilized for haemoglobin synthesis after the red cells die. Iron is transferred from macrophages to plasma transferrin and so to bone marrow erythroblasts. Iron absorption is normally just sufficient to make up for iron loss. The dashed line indicates ineffective erythropoiesis.
30 / Chapter 3: Hypochromic anaemias Hepcidin Hepcidin is a 25 amino acid polypeptide produced by liver cells. It is the major hormonal regulator of iron homeostasis (Fig.3.4a). It inhibits iron release from macrophages, from intestinal epithelial cells and from other cells by its interac- tion with the transmembrane iron exporter, ferroportin. It accelerates degradation of ferroportin protein by lysosomes and so directly inhibits iron export from ferroportin. Raised hepcidin levels therefore profoundly affect iron metabolism by reducing its absorption and its release from macrophages and hepatocytes. Control of hepcidin expression The synthesis of hepcidin in the liver is stimulated or inhibited by several factors. These include iron status, plasma cytokines, erythropoiesis and hypoxia (Fig. 3.4a). Control of hepcidin synthesis and so of iron status is best considered by comparing the control in iron overload when, except in genetic haemo- chromatosis, plasma hepcidin levels are high with that in iron deficiency when plasma hepcidin levels are low (Fig. 3.4c). In iron overload, diferric transferrin stimulates liver sinu- soidal cells to secrete bone morphogenetic proteins (BMPs) (Fig. 3.4b). Among them BMP2 controls basal hepcidin levels while BMP6 is especially increased in iron overload. The BMPs bind to their receptors (BMPRs) on the hepatic cell membrane. Diferric transferrin also binds to and stabilizes the transferrin receptor 2 (TFR2) on the cell membrane. Binding of BMPs to BMPRs results in the formation of a signalling complex consisting of BMPRs and three proteins TFR2, hemojuvelin (HJV) and HFE. This complex stimulates hepcidin synthesis via the signalling proteins SMADs, which transfer the message to the cell nucleus. Diferric transferrin also binds to TFR1 to provide the cell with iron. In iron deficiency there is little if any circulating diferric transferrin. This results in reduced synthesis of BMPs by the liver sinusoidal cells (Fig. 3.4c). Also in the absence of binding to diferric transferrin, TFR2 is degraded or shed from the cell membrane and HFE is deviated to bind to the unoccupied TFR1. Iron deficiency also stimulates the protease matriptase 2 (TMPRSS6) to cleave HJV from the cell surface. The result of all these reactions is failure to form the necessary complex to signal for hepcidin synthesis. Finally, iron deficiency stimulates a histone deacetylase which acts on the hepcidin locus to sup- press the transcription of hepcidin. Erythroblasts secrete erythroferrone and probably other proteins, which suppress BMP-mediated signalling for hepci- din secretion (Fig. 3.4a). This results in low plasma hepcidin levels in conditions with increased numbers of early erythro- blasts in the marrow, e.g. conditions of ineffective erythropoiesis, such as thalassaemia major, so iron absorption is inappro- priately increased despite the presence of iron overload due to transfusions. Hypoxia also suppresses hepcidin synthesis, whereas in inflammation interleukin 6 (IL-6) and other cytokines increase hepcidin synthesis (Fig. 3.4a). This results in lowering of plasma iron and helps to protect against infection by depriving bacteria of non-transferrin-bound iron in plasma. DMT1 Low iron Iron regulatory protein (IRP) 5′ AAAA mRNA stabilized IREs(5) IRE 3′ 5′ AAAA 3′ High iron ALAS Ferritin Coding region Coding region Translation blocked TFR1 Figure 3.3 Regulation of transferrin receptor 1 (TFR1), δ-aminolaevulinic acid synthase (ALA-S), divalent metal transporter 1 (DMT1) and ferritin expression by iron regulatory protein (IRP) sensing of intracellular iron levels. IRPs are able to bind to stem-loop structures called iron response elements (IREs). IRP binding to the IRE within the 3′ untranslated region of TFR1 and DMT1 leads to stabilization of the mRNA and increased protein synthesis, whereas IRP binding to the IRE within the 5′ untranslated region of ferritin and ALA-S mRNA reduces translation. IRPs can exist in two states: at times of high iron levels the IRP binds iron and exhibits a reduced affinity for the IREs, whereas when iron levels are low the binding of IRPs to IREs is increased. In this way, synthesis of TfR, ALAS, DMT1 and ferritin is coordinated to physiological requirements.
Chapter 3: Hypochromic anaemias / 31 Hepcidin SMADs (b) TFR1 HFE TFR2 Diferric transferrin Iron overload HJV BMPRs BMPs Hepatocyte Liver sinusoidal cell Hepcidin Histone deacetylase 3 (c) TFR1 HFE TFR2 Apo-transferrin Iron deficiency HJV BMPRs BMPs matriptase2 Hepatocyte Liver sinusoidal cell Duodenum FPN Transferrin saturation Hepcidin synthesis High transferrin saturation Inflammation IL6 Erythroblasts erythroferrone Hypoxia Erythropoietin Matriptase Inhibitor of hepciidin expression Low transferrin saturation Erythropoiesis Other tissues (a) Iron absorption Iron release from macrophage FPN Hepatocyte Figure 3.4 (a) Hepcidin reduces iron absorption and release from macrophages by stimulating degradation of ferroportin. Its synthesis is increased by diferric transferrin and by inflammation, but reduced by increased erythropoiesis and hypoxia. (b) The proposed mechanism by which the degree of transferrin saturation by iron in iron overload stimulates hepcidin synthesis (see also text). BMP synthesis by liver sinusoidal cells is stimulated by diferric transferrin, which also stabilizes TFR2 and by binding to TFR1 prevents HFE being deviated by attaching to TFR1. Binding of BMPs to BMPRs stimulates, in a complex with HFE, TFR2 and HJV on the hepatic cell membrane, hepcidin synthesis by signalling through SMAD proteins. (c) In iron deficiency low concentrations of diferric transferrin result in reduced BMP synthesis. Also lack of diferric iron to bind to TFR2 results in degradation or shedding of TFR2 from the cell membrane. Additionally, failure of diferric transferrin to bind to TFR1 results in HFE binding to TFR1. Iron deficiency also activates matriptase2, which cleaves HJV. The result of all these actions is failure to form the necessary complex to stimulate hepcidin synthesis. In addition in iron deficiency, a histone deacetylase 3 is activated and this suppresses transcription of hepcidin. BMP, bone morphogenetic protein; BMPR, bone morphogenetic protein receptor; HJV, hemojuvelin; TFR1 and 2, transferrin receptor 1 and 2. Source: (b and c) Courtesy of Professor Clara Camaschella.
32 / Chapter 3: Hypochromic anaemias Dietary iron Iron is present in food as ferric hydroxides and as ferric–protein and haem–protein complexes. Both the iron content and the proportion of iron absorbed differ from food to food; in general meat, in particular liver, is a better source than vege- tables, eggs or dairy foods. The average Western diet contains 10–15 mg iron daily, from which only 5–10% is normally absorbed. The proportion can be increased to 20–30% in iron deficiency or pregnancy (Table 3.2), but even in these situa- tions most dietary iron remains unabsorbed. Iron absorption Organic dietary iron is partly absorbed as haem and partly bro- ken down in the stomach and duodenum to inorganic iron. Absorption occurs through the duodenum. Haem is absorbed through a receptor on the apical membrane of the duodenal enterocyte. It is then broken down to release its iron. Inorganic iron absorption is favoured by factors such as acid and reducing agents that keep iron in the gut lumen in the Fe 2+ rather than the Fe 3+ state (Table 3.2). The protein DMT1 is involved in transfer of inorganic iron from the lumen of the gut across the enterocyte microvilli (Fig. 3.5). Ferroportin at the basolateral surface controls the exit of iron from the cell into portal plasma. The amount of iron absorbed is regulated according to the body’s needs by changing the levels of DMT1 and ferroportin. For DMT1 this occurs by the IRP/IRE binding mechanism (Fig. 3.3) and for ferroportin by hepcidin (Fig. 3.4a). Ferroportin is also present in liver, heart, kidney, brain and pla- centa, where it is important in exporting iron. Ferrireductase present at the enterocyte’s apical surface converts iron from the Fe 3+ to Fe 2+ state and another enzyme, hephaestin (ferrioxidase), converts Fe 2+ to Fe 3+ at the basal surface prior to its binding to transferrin. Iron requirements The amount of iron required each day to compensate for losses from the body and for growth varies with age and sex; it is highest in pregnancy, adolescent and menstruating females (Table 3.3). Therefore these groups are particularly likely to develop iron deficiency if there is any additional iron loss or prolonged reduced dietary intake. Table 3.2 Iron absorption. Factors favouring absorption Factors reducing absorption Inorganic iron Haem iron Ferrous form (Fe 2+ ) Ferric form (Fe 3+ ) Acids (hydrochloric acid, vitamin C) Alkalis – antacids, pancreatic secretions Solubilizing agents, e.g. sugars, amino acids Precipitating agents, e.g. phytates, phosphates, tea Reduced serum hepcidin e.g. iron deficiency Increased serum hepcidin e.g. inflammation Ineffective erythropoiesis (increased plasma erythroferrone) Decreased erythropoiesis (decreased plasma erythroferrone) Pregnancy Iron overload (acquired) Hereditary haemochromatosis Table 3.3 Estimated daily iron requirements. Units are mg/day. Urine, sweat, faeces Menses Pregnancy Growth Total Adult male 0.5–1 0.5–1 Postmenopausal female 0.5–1 0.5–1 Menstruating female* 0.5–1 0.5–1 1–2 Pregnant female* 0.5–1 1–2 1.5–3 Children (average) 0.5 0.6 1.1 Female (age 12–15)* 0.5–1 0.5–1 0.6 1.6–2.6 * These groups are more likely to develop iron deficiency.
Chapter 3: Hypochromic anaemias / 33 Iron deficiency Clinical features When iron deficiency is developing, the reticuloendothe- lial stores (haemosiderin and ferritin) become completely depleted before anaemia occurs (Fig.3.6). As the deficiency progresses, the individual may show the general symptoms and signs of anaemia (Chapter 2); and also painless glossitis, angular stomatitis, brittle, ridged or spoon nails (koilonychia; Fig. 3.7), hair loss and unusual dietary cravings (pica). Fatigue and depression are frequent and may occur even before anaemia is present. Oral or parenteral iron has been shown to reduce fatigue in iron-deficient (low serum ferritin) non-anaemic women. The cause of the epithelial cell changes may be related to reduction of tissue iron-containing enzymes. Neonatal iron deficiency is associated with cognitive and behavioural abnor- malities, while in children it can cause irritability, poor cogni- tive function and a decline in psychomotor development. Causes of iron deficiency In developed countries, chronic blood loss, especially uterine or from the gastrointestinal tract, is the dominant cause of iron deficiency (Table3.4) and dietary deficiency is rarely the sole cause. Five hundred millilitres of blood contain approx- imately 250 mg iron and, despite the increased absorption of iron at an early stage of iron deficiency, negative iron balance is usual in chronic blood loss. Ferritin Portal plasma Haem Haem oxygenase DMT-1 Ferrireductase Hepcidin Ferroportin Ferrioxidase Transferrin Mitochondrion Fe 3+ Fe 3+ Fe 2+ Figure 3.5 The regulation of iron absorption. Dietary ferric (Fe 3+ ) iron is reduced to Fe 2+ by ferrireductase and its entry to the enterocyte is through the divalent cation binder DMT-1. Its export into portal plasma is controlled by ferroportin. It is oxidized by hephaestin (ferrioxidase) before binding to transferrin in plasma. Haem is absorbed after binding to its receptor protein. Red cell iron (peripheral film and indices) Normal Normal ++ Normal Latent iron deficiency Iron deficiency anaemia Hypochromic, microcytic MCV↓ MCH↓ Iron stores (bone marrow macrophage iron) 0 0 Figure 3.6 The development of iron deficiency anaemia. Reticuloendothelial (macrophage) stores are lost completely before anaemia develops. MCH, mean corpuscular haemoglobin; MCV, mean corpuscular volume.
34 / Chapter 3: Hypochromic anaemias Increased demands for iron during infancy, adolescence, pregnancy, lactation and in menstruating women account for the high risk of iron deficiency anaemia in these particular clinical groups. Newborn infants have a store of iron derived from delayed clamping of the cord and the breakdown of excess red cells. From 3 to 6 months, there is a tendency for a negative iron balance because of growth. From 6 months, supplemented formula milk and mixed feeding, particularly with iron- fortified foods, prevent iron deficiency. In pregnancy increased iron is needed for an increased maternal red cell mass requiring approximately 600 mg iron, transfer of 300 mg of iron to the foetus and because of blood loss at delivery (Chapter 34). Menorrhagia (a loss of 80 mL or more of blood at each cycle) is difficult to assess clinically, although the loss of clots, the use of large numbers of pads or tampons or prolonged periods all suggest excessive loss. It takes about 8 years for an adult male starting with nor- mal iron stores to develop iron deficiency anaemia solely as a result of a poor diet or malabsorption resulting in no iron intake. In developed countries inadequate intake or malabsorp- tion is only rarely the sole cause of iron deficiency anaemia. Gluten-induced enteropathy, partial or total gastrectomy and atrophic gastritis (often auto-immune and with Helicobacter pylori infection) may, however, predispose to iron deficiency. Table 3.4 Causes of iron deficiency. Chronic blood loss Uterine Gastrointestinal, e.g. peptic ulcer; oesophageal varices; aspirin (or other non-steroidal anti-inflammatory drugs) ingestion; gastrectomy; carcinoma of the stomach, caecum, colon or rectum; hookworm; schistosomiasis; angiodysplasia; inflammatory bowel disease; piles; diverticulosis Rarely, haematuria, haemoglobinuria, pulmonary haemosiderosis, self-inflicted blood loss Increased demands (see also Table 3.3) Prematurity Growth Pregnancy Erythropoietin therapy Malabsorption Gluten-induced enteropathy, gastrectomy, autoimmune gastritis, bariatric surgery, Helicobacter infection Poor diet A major factor in many developing countries, but rarely the sole cause in developed countries (a) (b) Figure 3.7 Iron deficiency anaemia. (a) Koilonychia: typical ‘spoon’ nails. (b) Angular cheilosis: fissuring and ulceration of the corner of the mouth.
Chapter 3: Hypochromic anaemias / 35 In developing countries, iron deficiency may occur as a result of a life-long poor diet, consisting mainly of cereals and veg- etables. Hookworm or schistosomiasis may aggravate iron deficiency, as may repeated pregnancies, growth in children and menorrhagia in young females. Laboratory findings These are summarized and contrasted with those in other hypochromic anaemias in Table 3.7. Red cell indices and blood film Even before anaemia occurs, the red cell indices fall, and they fall progressively as the anaemia becomes more severe. The blood film usually shows hypochromic, microcytic red cells with occa- sional target cells and pencil-shaped poikilocytes (Fig. 3.8). Rarely the red cell indices may be normal. The reticulocyte count is low in relation to the degree of anaemia. When iron deficiency is associated with severe folate or vitamin B 12 deficiency, a ‘dimorphic’ film occurs with a dual population of red cells of which one is macrocytic and the other microcytic and hypochro- mic; the indices may be normal. A dimorphic blood film is also seen in patients with iron deficiency anaemia who have received recent iron therapy and produced a population of new haemo- globinized normal-sized red cells (Fig. 3.9) and when the patient has been transfused. The platelet count is often moderately raised in iron deficiency, particularly when haemorrhage is con- tinuing. Low iron promotes megakaryocyte commitment of megakaryocytic-erythroid bone marrow progenitors. Bone marrow iron Bone marrow examination is not needed to assess iron stores except in complicated cases. In iron deficiency anaemia, there is a complete absence of iron from stores (macrophages) and from developing erythroblasts (Fig. 3.10). The erythroblasts are small and have a ragged cytoplasm. Serum iron, total iron-binding capacity (TIBC) and transferrin saturation (TSAT) The serum iron falls but this assessment alone is not useful. The total iron-binding capacity (TIBC) (or transferrin level) rises so that the TSAT (the ratio between serum iron and TIBC or transferrin expressed as a percentage) is <16% though some take <20% as abnormal (Fig. 3.11). This contrasts both with the anaemia of chronic disease (see below), when the serum iron and the TIBC are both reduced, and with other hypochro- mic anaemias where the serum iron is normal or even raised. Because of the diurnal fluctuations in serum iron, the TSAT may also fluctuate. Serum ferritin A small fraction of body ferritin circulates in the serum, the concentration being related to tissue, particularly reticuloen- dothelial, iron stores. The normal range in men is higher than in women (Fig. 3.11). In iron deficiency anaemia, the serum ferritin is very low <15 μg/L. A raised serum ferritin indicates iron overload, excess release of ferritin from damaged tissues or an acute phase response, e.g. in inflammation. The serum ferritin is normal or raised in the anaemia of chronic disorders so it cannot exclude iron deficiency in this setting. If an active inflammatory state is present, a serum ferritin <150 ug/L suggests iron deficiency may also be present. i Other tests Plasma soluble transferrin receptors and red cell zinc pro- toporphyrin levels are raised in iron deficiency but neither are sufficiently specific or sensitive to be recommended for routine use. Assays of serum hepcidin have been developed and are likely to become available in the routine clinical laboratory. Figure 3.8 The peripheral blood film in severe iron deficiency anaemia. The cells are microcytic and hypochromic with occasional target cells. Figure 3.9 Dimorphic blood film in iron deficiency anaemia responding to iron therapy. Two populations of red cells are present: one microcytic and hypochromic, the other normocytic and well haemoglobinized.
36 / Chapter 3: Hypochromic anaemias Investigation of the cause of iron deficiency (Fig. 3.12) In premenopausal women, menorrhagia and/or repeated preg- nancies are the usual causes. If these are not present, other causes must be sought. In some patients with menorrhagia, a clotting or platelet abnormality, e.g. von Willebrand disease is present. In poor countries, the combination of prolonged inadequate intake of iron with expansion of blood volume with growth, and the onset of menstruation and repeated pregnancies in young women, is the major cause of iron deficiency in childhood, adolescence and fertile adult females. In men and postmenopausal women, gastrointestinal blood loss is the main cause of iron deficiency and the exact site is sought from the clinical history, physical and rec- tal examination, by occult blood tests, and by appropriate use of Normal Iron deficiency Anaemia of chronic disease Iron overload 0 30 (μmol/L) 60 90 Serum iron UIBC 0 200 100 300 Serum ferritin (μg/L) 1000 10 000 M F Figure 3.11 The serum iron, unsaturated serum iron-binding capacity (UIBC) and serum ferritin in normal subjects and in those with iron deficiency, anaemia of chronic disease and iron overload. The total iron-binding capacity (TIBC) is made up of the serum iron and the UIBC. In some laboratories, the transferrin content of serum is measured directly by immunodiffusion, rather than by its ability to bind iron, and is expressed in g/L. Normal serum contains 2–4 g/L transferrin (1 g/L transferrin = 20 μmol/L binding capacity). Normal ranges for serum iron are 10–30 μmol/L; for TIBC, 40–75 μmol/L; for serum ferritin, male, 40–340 μg/L; female, 14–150 μg/L. (b) (a) Figure 3.10 Bone marrow iron assessed by Perls’ stain. (a) Normal iron stores indicated by blue staining in the macrophages. Inset: normal siderotic granule in erythroblast. (b) Absence of blue staining (absence of haemosiderin) in iron deficiency. Inset: absence of siderotic granules in erythroblasts.
Chapter 3: Hypochromic anaemias / 37 upper and lower gastrointestinal endoscopy and/or radiology, e.g. computed tomography (CT) of the pneumocolon, or virtual colonoscopy using the 3D colon system (Figs. 3.12 and 3.13). In difficult cases, a camera in a capsule can be swallowed, which relays pictures of the gastrointestinal tract electronically. Tests for parietal cell antibodies, Helicobacter infection and serum gastrin level may help to diagnose autoimmune gastritis. Tests for transglutaminase antibodies and duodenal biopsy to look for gluten-induced enteropathy may be needed. Hookworm and schistosomiasis ova are sought in stools of subjects from areas where these infestations are endemic. Serological tests for schisto- somiasis can also be performed. Rarely, a coeliac axis angiogram is needed to demonstrate angiodysplasia. If gastrointestinal blood loss is excluded, loss of iron in the urine as haematuria or haemosiderinuria (resulting from chronic intravascular haemolysis) is considered. A normal chest X-ray excludes the rare condition of pulmonary haemosiderosis. Rarely, patients bleed themselves, producing iron deficiency. Treatment The underlying cause is treated as far as possible. In addition, iron is given to correct the anaemia and replenish iron stores. Oral iron The best preparations are ferrous sulphate (200 mg) and fer- rous fumarate (210 mg), which are cheap and contain 67 mg iron in each tablet. Iron is best absorbed if they are taken on an empty stomach. In the past doses two or three times daily were used to overcome limited absorption. Once daily doses are now advocated based on the rise in plasma hepcidin level (which blocks further iron absorption) within hours of a single oral dose of iron. For those without anaemia or only mildly anae- mic, alternate day dosing may be used. For those with more severe anaemia, daily doses or even intravenous iron are needed to correct the anaemia more quickly. If side effects occur, e.g. nausea, abdominal pain, constipation or diarrhoea, these can be reduced by giving iron with food or by using a preparation with a lower iron content, e.g. ferrous gluconate, which con- tains only 37 mg per 300 mg tablet. An elixir of iron is available for children. Slow-release preparations should not be used. Oral iron therapy should be given for long enough both to correct the anaemia and to replenish body iron stores, which usually means for at least up to 6 months. The haemo- globin should rise at the rate of approximately 20 g/L every 3 weeks. Failure of response to oral iron has several possible causes (Table 3.5). These should all be considered before parenteral Suspicion Diagnosis Investigation of cause Treatment HYPOCHROMIC MICROCYTIC ANAEMIA Low serum iron and raised TIBC Low serum ferritin Female Male or female Menorrhagia GI blood loss Occult blood test Upper and lower GI endoscopy Repeated pregnancies Investigation of other causes (see Table 3.4) 1. Treat cause 2. Oral iron, e.g., ferrous sulphate to correct anaemia and replenish stores (Parenteral iron may be needed) Figure 3.12 Investigation and management of iron deficiency anaemia. GI, gastrointestinal; TIBC, total iron-binding capacity. Figure 3.13 Virtual colonoscopy showing carcinoma of colon causing colonic obstruction and iron deficiency. Table 3.5 Failure of response to oral iron. Continuing haemorrhage Failure to take tablets Wrong diagnosis – especially thalassaemia trait, sideroblastic anaemia, IRIDA Mixed deficiency – associated folate or vitamin B 12 deficiency Another cause for anaemia e.g. malignancy, inflammation Malabsorption – see Table 3.4 Use of slow-release preparation IRIDA, iron-refractory iron deficiency anaemia.
38 / Chapter 3: Hypochromic anaemias iron is used. Iron fortification of the diet in infants in Africa reduces the incidence of anaemia, but increases susceptibility to malaria. Parenteral iron Many different preparations are available with varying licensing arrangements in different countries (Table 3.6). The dose may be calculated according to body weight and degree of anaemia, but it is more usual in order not to waste costly drugs to give 1000 mg as a single dose to moderately anaemic patients and 1500 mg divided into two doses, given at least 48 hours apart, to those more severely anaemic or of large body mass. All the preparations contain ferric (usually ferric hydroxide) iron. Iron dextran can be given in small doses by slow intravenous injection or by infusion as a total dose in one day. Ferric carboxymaltose and ferric iso- maltoside may be given as a total dose in one day by intravenous infusion or by slow intravenous injections. Ferumoxytol is given only by intravenous infusion. Ferric hydroxide–sucrose, which releases iron from its sugar more rapidly than the other prepara- tions, is administered by slow intravenous injection or infusion, to a maximum of 200 mg iron in each dose. Rarely there may be hypersensitivity or anaphylactoid reac- tions to parenteral iron, especially in those with a previous reaction, multiple drug allergies and severe atopy. If the reac- tion is severe, it is treated with intravenous hydrocortisone and possibly adrenaline. Parenteral iron is given when there are high iron require- ments, as in gastrointestinal bleeding, severe menorrhagia, middle or late pregnancy (avoided in the first trimester), chronic haemodialysis and chronic renal failure with erythro- poietin therapy, post-operative after major surgery. It is also given when oral iron is ineffective, e.g. iron malabsorption (Table 3.4), if IRIDA (see below) is present, or when oral iron causes intolerable side effects or is impractical, e.g. active inflammatory bowel disease. The haematological response to parenteral iron is no faster than to adequate dosage of oral iron, but the iron stores are replenished faster. Intravenous iron may increase functional capacity and quality of life in some patients with congestive heart failure, even in the absence of anaemia (see p. xxx). It may also be effective in restless leg syndrome. Early trials show it can reduce blood transfusion needs in chemotherapy-induced anaemia in cancer patients. Anaemia of chronic disease (inflammation) One of the most common anaemias occurs in patients with a variety of chronic inflammatory and malignant diseases (Table 3.7). With iron deficiency it accounts for about two- thirds of the world’s anaemias. The characteristic features are: 1 Normochromic, normocytic or mildly hypochromic (MCV rarely <75 fL) indices and red cell morphology. 2 Mild and non-progressive anaemia (haemoglobin rarely <90 g/L) – the severity being related to the severity of the underlying disease. Table 3.7 Causes of the anaemia of chronic disease. Chronic inflammatory diseases Infections, e.g. pulmonary abscess, tuberculosis, osteomyelitis, pneumonia, bacterial endocarditis Non-infectious, e.g. rheumatoid arthritis, systemic lupus erythematosus and other connective tissue diseases, sarcoidosis, inflammatory bowel disease Other chronic disorders Congestive heart failure Chronic pulmonary disease Chronic renal disease Obesity Anaemia in the elderly Anaemia in critical illness (accelerated course) Malignant diseases Carcinoma, lymphoma, sarcoma Table 3.6 Intravenous iron preparations. Trade name Cosmofer (Europe) INFeD (USA) Ferinject (Europe) Injectafer (USA) Feraheme Monofer, Diafer (Europe) Monoferric (USA) Ferrlecit Venofer (Europe only) Carbohydrate Low molecular weight dextran Carboxymaltose Ferumoxytol Derisomaltoside Gluconate in sucrose solution Sucrose Vial 50 mg/μL 50 mg/ml 30 mg/mL 100 mg/mL or 50 mg/mL 5 mL (12.5 mg/mL) 20 mg/mL Total dose Yes Yes Yes Yes No No Test dose required Yes No No No No No Infusion time 1 h 15 min 15 min 20 min 60 min 15 min Slow intravenous injection Yes Yes No Yes Yes Yes
Chapter 3: Hypochromic anaemias / 39 3 Both the serum iron and TIBC are reduced. 4 The serum ferritin and hepcidin levels are normal or raised. 5 Bone marrow storage (reticuloendothelial) iron is normal, but erythroblast iron is reduced (Table 3.8). 6 Raised plasma levels of hepcidin and inflammation-induced cytokines block intestinal iron absorption and cause iron retention in the reticuloendothelial cells. 7 Shortened reduced red cell life span, suppressed erythro- poietin response to anaemia and inhibited erythroid cell differentiation. The pathogenesis of this anaemia is related to decreased release of iron from macrophages to plasma, because of raised serum hepcidin levels stimulated by IL-6 and IL-1β and lipopolysaccharide, a mechanism to protect the host against bacteria which need iron to multiply. There is also an inade- quate erythropoietin response to anaemia caused by the effects of cytokines such as IL-1, IL-6, IL-10 interferon-gamma and tumour necrosis factor. These cytokines directly damage red cells whose life span is also reduced both by activation of mac- rophages and by deposition of antibody and complement on their surface. Interferon- γ further impairs haemopoiesis by binding to thrombopoietin, inhibiting thrombopoietin bind- ing to its receptor on marrow cells. The anaemia is corrected by successful treatment of the underlying disease. It does not respond to iron therapy but diagnosis of the anaemia when there is coexisting iron defi- ciency may be difficult. If the serum ferritin is <100 ug/L or saturation of the TIBC <20%, co-existing iron deficiency is suspected. The combination of iron therapy and erythropoie- tin injections improve the anaemia in some cases. In many con- ditions, this anaemia is complicated by anaemia resulting from other causes, e.g. vitamin B 12 or folate deficiency, renal failure, bone marrow failure, hypersplenism, endocrine abnormality or leucoerythroblastic anaemia. These are discussed in Chapter 32. Measurement of serum hepcidin and drugs to inhibit hepcidin, both being developed, would be major advances in the diagnosis and treatment of the anaemia of chronic disease. Iron refractory iron deficiency anaemia (IRIDA) This rare autosomal recessive syndrome presents with an hypochromic microcytic anaemia. The serum iron is low with <5% saturation of the iron binding capacity (Table 3.8). It is caused by inherited mutations of matriptase 2, which allow uninhibited hepcidin secretion, or even more rarely mutations of DMT1 genes (Figs. 3.4 and 3.5). There may be a haematological response to intravenous but usually not to oral iron. Sideroblastic anaemia This is an uncommon refractory anaemia defined by the presence of many pathological ring sideroblasts in the bone marrow (Fig.3.14). These are erythroblasts containing numerous iron granules arranged in a ring or collar around the nucleus, instead of the one or two randomly distributed iron granules in normal erythroblasts (Fig. 3.10). There is also usu- ally erythroid hyperplasia with ineffective erythropoiesis. Primary sideroblastic anaemia is diagnosed when 15% or more of marrow erythroblasts are ring sideroblasts but in the pres- ence of the SF3B1mutation as few as 5% ring sideroblasts in Table 3.8 Laboratory diagnosis of a hypochromic anaemia. Iron deficiency Chronic disease Thalassaemia trait (α or β) Sideroblastic anaemia IRIDA MCV/ MCH Reduced in relation to severity of anaemia Normal or mild reduction Reduced; low for degree of anaemia Usually low in congenital type but MCV usually raised in acquired type Reduced in relation to severity of anaemia Serum iron Reduced Reduced Normal Raised Reduced TIBC Raised Reduced Normal Normal Reduced Serum ferritin Reduced Normal or raised Normal Raised Raised Bone marrow iron stores Absent Present Present Present Present Erythroblast iron Absent Absent Present Ring forms Absent Haemoglobin electrophoresis Normal Normal Hb A 2 raised in β form Normal Normal Hb, haemoglobin; IRIDA, non-refractory iron deficiency anaemia; MCH, mean corpuscular haemoglobin; MCV, mean corpuscular volume; TIBC, total iron- binding capacity.
40 / Chapter 3: Hypochromic anaemias patients with myelodysplasia is still classified as a ‘primary’ sideroblastic anaemia (see below). Ring sideroblasts can also be found at lower numbers in a variety of other anaemias. Sideroblastic anaemia is classified into different types (Table3.9), the common link is a defect in haem synthesis. In the hereditary forms, the anaemia is usually markedly hypochromic and microcytic but some types show normocytic or macrocytic red cells. The most common mutations are in the ALAS gene, which is on the X chromosome. Pyridoxal-6- phosphate is a coenzyme for ALAS. Other rare types include an X-linked mitochondrial disease with spino-cerebellar degeneration and ataxia, mutations in other mitochondrial genes with the triad of anaemia, deafness and diabetes or Pearson syndrome (when there is also pancreatic insuffi- ciency). Thiamine-responsive anaemia with both megaloblas- tic and sideroblastic erythropoiesis, accompanied by deafness and diabetes is due to mutations of the gene which codes for a thiamine transporter. Various congenital malformations of the limbs, face, nervous system, heart and other organs may be present in these rare syndromes. The most frequent of the acquired sideroblastic anaemias is refractory anaemia with ring sideroblasts with the SF3B1 mutation, a subtype of myelodysplasia (Chapter 16). Acquired reversible forms may be due to alcohol, lead and drugs, e.g. isoniazid. In some patients, particularly with the hereditary ALAS mutated type, there is a response to pyridoxine therapy. High doses of thiamine (vitamin B1) improve those with the thiamine transporter gene mutation. Luspatercept (Chapters 7, 16) can be tried. Folate deficiency may occur and folic acid therapy may also be tried. Other treatments, e.g. erythropoietin and luspa- tercept, are effective in some patients with the myelodysplasia form (Chapter 16). In many severe cases, however, repeated blood transfusions are the only method of maintaining a satisfac- tory haemoglobin concentration, and transfusional iron overload requiring iron chelation therapy becomes a major problem. Lead poisoning Lead inhibits both haem and globin synthesis at a number of points. In addition, it interferes with the breakdown of RNA by inhibiting the enzyme pyrimidine 5′ nucleotidase, causing accumulation of denatured RNA in red cells, the RNA giving Figure 3.14 Ring sideroblasts with a perinuclear ring of iron granules in sideroblastic anaemia. Table 3.9 Classification of sideroblastic anaemia. Hereditary X chromosome-linked ALAS mutation usually occurs in males, transmitted by females; also occurs rarely in females X-linked mitochondrial gene mutation with spino-cerebellar ataxia. Other rare autosomal types usually involving mitochondrial proteins or thiamine phosphorylation; deafness and diabetes often present Acquired Primary Myelodysplasia (refractory anaemia with ring sideroblasts, as low as 5% with SF3B1 mutation; see Chapter 16). N.B. Ring sideroblast formation (<15% of erythroblasts) may also occur in the bone marrow inother malignant diseases of the marrow, e.g. other types of myelodysplasia, myelofibrosis, myeloid leukaemia, myeloma, drugs, e.g. anti-tuberculosis (isoniazid, cycloserine), alcohol, lead other benign conditions, e.g. haemolytic anaemia, megaloblastic anaemia, rheumatoid arthritis ALAS, δ-aminolaevulinic acid synthase.
Chapter 3: Hypochromic anaemias / 41 an appearance called basophilic stippling on the ordinary (Romanowsky) stain (Fig. 2.17). The anaemia may be hypochro- mic or predominantly haemolytic, and the bone marrow may show ring sideroblasts. Free erythrocyte protoporphyrin is raised. Differential diagnosis of hypochromic anaemia Table 3.8 lists the laboratory investigations that may be neces- sary. The clinical history is particularly important, as the source of the haemorrhage leading to iron deficiency or the presence of an inflammatory or other chronic disease may be revealed. The ethnic group and the family history may suggest a possible diagnosis of thalassaemia or other genetic defect of haemoglo- bin. Physical examination may also be helpful in determining a site of haemorrhage, features of a chronic inflammatory or malignant disease, koilonychia or, in some haemoglobinopa- thies, an enlarged spleen or bony deformities. In thalassaemia trait the red cells tend to be very small, often with an MCV of 70 fL or less, even when anaemia is mild or absent; the red cell count is usually over 5.5 × 10 12 /L. Conversely, in iron deficiency anaemia the indices fall progressively with the degree of anaemia and when anaemia is mild the indices may be normal or only just reduced, e.g. MCV 75–80 fL. In the anae- mia of chronic disorders, the indices are also not markedly low, an MCV in the range 75–82 fL being usual. It is usual to measure serum iron and TIBC, and/or serum ferritin to confirm iron deficiency. Haemoglobin high- performance liquid chromatography (HPLC) or electrophore- sis with an estimation of Hb A 2 and Hb F is carried out in all patients suspected of thalassaemia or other genetic defect of haemoglobin, because of the family history, ethnic group, red cell indices and blood film. Iron deficiency or the anaemia of chronic disorders may also occur in these subjects. β- Thalassaemia trait is characterized by a raised Hb A 2 above 3.5%, but in α-thalassaemia trait there is no abnormality on simple haemoglobin studies, so the diagnosis is usually made by exclusion of all other causes of hypochromic red cells and by the presence of a red cell count >5.5 × 10 12 /L. DNA studies confirm the diagnosis. Bone marrow examination is essential if a diagnosis of side- roblastic anaemia is suspected, but is not usually needed in diagnosis of the other hypochromic anaemias. ■ Iron is present in the body in haemoglobin, myoglobin, haemosiderin and ferritin, and in iron-containing enzymes. Transferrin is the main transport protein in blood. ■ Hepcidin is the main regulator of iron absorption and iron release from macrophages and other cells. ■ Iron metabolism is regulated according to iron status by intracellular iron regulatory proteins and by control of hepcidin synthesis. Hepcidin synthesis is also affected by erythroferrone secreted by erythroblasts and by inflammation. ■ Iron deficiency is the most common cause of anaemia throughout the world. The red cells are hypochromic and microcytic. The serum ferritin, serum iron and saturation of the iron-binding capacity are reduced. ■ In Western countries, it is usually caused by haemorrhage from the gastrointestinal or the female genital tract. ■ Prolonged reduced dietary intake may cause the anaemia, particularly in developing countries, where hookworm and schistosomiasis may also be important causes of blood loss. ■ It is treated by oral or less frequently parenteral iron to correct the anaemia and to restore body stores of iron, and by treating, as far as possible, the underlying cause. ■ Other frequent causes of a hypochromic, microcytic anaemia are the anaemia of chronic disease, which occurs in patients with chronic inflammatory or malignant diseases, and α- or β-thalassaemia. Less common causes include sideroblastic anaemia (some cases), iron refractory iron deficiency anaemia and lead poisoning. ■ Sideroblastic anaemias characterized by ring sideroblasts in the marrow are rare. They may be inherited or acquired; the most common subtype is myelodysplasia with the SF3B1 mutation. SUMMARY Now visit www.wiley.com/go/haematology9e to test yourself on this chapter.
Chapter 4: Iron overload / 43 There is no physiological mechanism for eliminating excess iron from the body. Iron absorption is carefully regulated to avoid accumulation of excess iron. Iron overload (haemosi- derosis) occurs in genetic disorders associated with inap- propriately increased iron absorption (haemochromatosis), or in patients with severe chronic anaemias, especially those who receive regular blood transfusions. Excessive iron dep- osition in tissues may result in serious damage particularly to the heart, liver and endocrine organs. The causes of iron overload are listed in Table 4.1 and causes of hereditary haemo- chromatosis in Table 4.2. Assessment of iron status and organ function The tests to assess iron overload and the degree of damage caused by iron are listed in Table 4.3. The serum ferritin is the most widely used test to assess iron overload and to monitor its treatment, although inflammatory and other diseases may increase serum ferritin levels even in the absence of iron over- load. The percentage saturation of the iron-binding capacity (transferrin) is also valuable. Serum non-transferrin-bound (NTBI) iron, also known as labile plasma iron, is a toxic form of iron that occurs in severe transfusional iron overload but tests for NTBI are not widely available. Aceruloplasminaemia is another rare genetic cause of iron loading. Table 4.1 The causes of iron overload. Increased iron absorption Hereditary (genetic) haemochromatosis Ineffective erythropoiesis, e.g. non- transfusion-dependent thalassaemia or myelodysplastic syndromes Chronic liver disease Increased iron intake African siderosis (dietary and genetic components) Repeated red cell transfusions Transfusion siderosis Table 4.2 Genetic causes of haemochromatosis and of hyperferritinaemia. Type Inheritance Clinical condition Gene defect 1 AR Classical hereditary haemochromatosis HFE 2 AR Juvenile haemochromatosis Hemojuvelin (HJV) Hepcidin (HAMP) 3 AR Hereditary haemochromatosis Transferrin receptor 2 (TFR2) 4a AD RE iron loading Ferroportin (SLC11A3) 4b AD Parenchymal iron loading Ferroportin (SCL11A3) (mutation at the hepcidin binding site) 5 AD Hereditary hyperferritinaemia – cataract syndrome (no iron deposition) Ferritin light chain (FTL) AD, autosomal dominant; AR, autosomal recessive; RE, reticuloendothelial. Table 4.3 Assessment of iron overload. Assessment of iron stores Serum ferritin Serum iron and percentage saturation of iron-binding capacity (transferrin) Serum non-transferrin-bound iron (NTBI) Bone marrow biopsy (Perls’ stain) for iron within reticuloendothelial(RE) stores Liver biopsy (parenchymal and RE stores) Liver CT scan or MRI (T 2 * or Ferriscan technique) Cardiac MRI (gated T 2 * technique) Pancreas and pituitary MRI (not widely available) Annual transfusional calculated iron loading Assessment of tissue damage caused by iron overload Cardiac Clinical; chest X-ray; ECG (24-h monitor); echocardiography (ECHO) to assess left and right ventricular ejection fraction at rest and with stress Liver Liver function tests, alpha-fetoprotein; liver ultrasound; MRI, fibroscan Endocrine and bone Clinical examination (including for growth and sexual development); oral glucose tolerance test; thyroid, parathyroid, gonadal, adrenal function and growth hormone assays; radiology for bone age; isotopic bone density study, vitamin D Musculoskeletal Hand X-rays with assessment of metacarpophalangeal joints CT, computed tomography; ECG, electrocardiography; MRI, magnetic resonance imaging.
44 / Chapter 4: Iron overload Liver biopsy with staining for iron (Fig. 4.1) and chemical analysis of iron content is useful for assessment of both paren- chymal iron (within hepatic cells) and reticuloendothelial iron (within Kupffer cells). Magnetic resonance imaging (MRI) using the T 2 * technique is the best non-invasive guide to liver and cardiac iron. A commercial Ferriscan MRI tech- nique is widely used for measuring liver iron. Liver biopsy also allows estimation of the degree of fibrosis. Fibroscan (transient elastography) is a non-invasive ultrasound method of assessing liver fibrosis from any cause, including iron over- load. Serum alpha-fetoprotein and liver ultrasound are used for serial screening for hepatocellular carcinoma in patients with known iron overload, previous hepatitis C or liver fibro- sis typically on an annual basis. Monitoring for cardiac iron overload and of cardiac function are discussed on page xx under the heading of transfusional iron overload where car- diac disease is a major complication. Other complications of transfusional iron overload including liver, endocrine and bone disease are also discussed further under the heading thalassaemia major in Chapter 7. Hereditary (genetic, primary) haemochromatosis Hereditary haemochromatosis is a group of diseases in which there is from birth excessive absorption of iron from the gastro- intestinal tract leading to iron overload of the parenchymal cells, dominantly of the liver (Fig. 4.1). At later stages endo- crine and heart complications occur. In contrast to transfu- sional iron overload, the macrophages are not iron overloaded. Most patients are homozygous for a missense mutation (C282Y) in the HFE gene, which leads to insertion of a tyrosine residue rather than cysteine in the mature protein. This allele has the highest prevalence (approximately 1 in 10) within populations of Northern European origin. However, gene penetrance is low; only a small proportion of this ethnic group who are homozygous for the mutation (about 1 in 300) present with clinical features of the disease. Affected individu- als usually show a serum ferritin greater than 1000 μg/L. A second mutation H63D resulting in a histidine to aspar- tic acid substitution is found with the C282Y mutation in approximately 5% of patients with genetic haemochromatosis. Homozygotes for the H63D mutation usually do not have the disease. The H63D mutation has a broader global distribution than C282Y. HFE is involved in regulation of hepcidin synthesis (Fig. 3.4). The C282Y mutation causes low plasma levels of hepcidin and so high levels of ferroportin in enterocytes, mac- rophages and other cells. Iron absorption and iron release from macrophages is therefore increased. Iron overload develops over decades and damages parenchymal cells, so that patients may present in adult life with hepatic disease (fibrosis, cirrho- sis, hepatocellular carcinoma), endocrine disturbances (diabe- tes mellitus, hypothyroidism or impotence) or melanin skin pigmentation (Fig. 4.2). In some severe cases, there is cardiac failure or arrhythmia but these are more dominant features of transfusional iron overload. Homozygosity for the HFE muta- tion also underlies an arthropathy due to calcium pyrophos- phate deposition and not related to the degree of iron overload. Most commonly affected are the second and third metacarpo- phalangeal joints. Other non- HFE genetic factors as well as dietary iron intake, alcohol consumption, pregnancies, menstrual blood loss, other blood loss, e.g. gastrointestinal bleeding or blood donation affect the phenotype and time of presentation of the Figure 4.1 Liver biopsy in genetic haemochromatosis. Iron loading of hepatic parenchymal cells (Perls’ stain). Source: Courtesy of Professor A.P. Dhillon.
Chapter 4: Iron overload / 45 disease. Regular blood donors on average present several years later than individuals who never donate blood. The initial clinical presentation is often with non-specific symptoms such as fatigue, loss of libido or arthralgias. Diagnosis is suspected by the presence of increased levels of serum iron, serum transferrin saturation and ferritin. The diagnosis is con- firmed by testing for the HFE mutation, but a negative result does not exclude the diagnosis, since mutations of other genes may cause a similar disease phenotype (Table 4.2 and below). Liver biopsy is useful to quantify the degree of iron overload and assess liver damage, but is associated with risks, including bleeding. MRI can be used instead to measure liver and cardiac iron. Treatment is with regular venesection, initially at 1–2- week intervals, each unit of blood removing 200–250 mg of iron. Organ function may improve and liver fibrosis resolve, but the arthropathy does not respond to iron removal and response of the endocrine damage is variable. There are differ- ences of opinion as to whether patients without evidence of organ dysfunction due to iron overload should be treated. Venesections are generally started when the serum ferritin is raised to over 1000 μg/L but some recommend starting if the serum ferritin is >300 μg/L in males or >200 μg/L in females. Venesection is monitored by serum ferritin and the aim is to restore this to normal, <100 μg/L, and then keep it there with further less frequent venesections, e.g. a few times per year. Some patients elect to become blood donors, but some blood banks and regulatory agencies do not allow blood donation from patients with haemochromatosis. Those with abnormal liver tests are excluded from donating due to the inability to exclude viral causes of hepatic enzyme elevation. Rarer forms of genetic haemochromatosis are caused by mutations in genes for other iron regulatory proteins, includ- ing hemojuvelin, hepcidin and transferrin receptor 2. All (types II and III; Table 4.2) are associated, like homozygous HFE disease, with low plasma levels of hepcidin. They often present below the age of 30 years as severe (especially for Type II) iron overload with cardiomyopathy in children, adolescents or young adults. Hypogonadism is another particular feature, whereas arthropathy is absent. Genetic iron overload in Asian populations is usually due to these mutations rather than mutation of HFE. On the other hand, ferroportin gene mutations (type IV disease) usually cause reticuloendothelial but not parenchymal cell iron overload. They do rarely cause parenchymal overload if the mutations in the ferroportin gene are at the hepcidin binding site. Aceruloplasminaemia is another rare genetic cause of iron loading. Mutations of the ferritin light chain gene (type V disease) cause a raised monoclonal serum ferritin with cata- racts resulting from ferritin deposition in the eye, but no other tissue iron overload. African iron overload This occurs in sub-Saharan Africa through a combination of increased iron absorption due to a genetic defect, possibly in the ferroportin gene, and consumption of beverages, especially beer, with a high iron content due to the use of iron brewing or cooking pots. Non-transfusion-dependent thalassaemia (thalassaemia intermedia) Moderately severe forms of thalassaemia may lead to increased iron levels even in patients who do not need regular blood transfusions (Chapter 7). This is due to increased iron absorp- tion due to raised plasma erythroferrone. This leads to increased levels of iron in the liver. The heart is spared. Blood transfu- sions at times of increased anaemia, e.g. with intercurrent infections, may increase the iron burden. Iron chelation is indi- cated if the liver iron concentration is above 5 mg/g dry weight or when the serum ferritin reaches 800 μg/L or when the iron leads to organ damage (see also Chapter 7). Other haemolytic anaemias including pyruvate kinase deficiency may also lead to iron loading due to low serum hepcidin levels caused by raised plasma erythroferrone due to ineffective erythropoiesis. Transfusional iron overload This develops in patients with chronic severe anaemia not due to haemorrhage who need regular blood transfusions. Each 500 mL of transfused blood contains 200–250 mg iron so iron overload is inevitable unless iron chelation therapy is given (Table4.4). To make matters worse, iron absorption from food is increased in β-thalassaemia major and other anaemias secondary to ineffective erythropoiesis despite tissue iron overload. This is due to release from early erythro- blasts of erythroferrone and of other proteins that inhibit hep- cidin synthesis (Fig. 3.4). Non-transferrin-bound iron may appear in plasma when transferrin is >70% saturated. It causes widespread iron deposition in parenchymal tissues. Figure 4.2 Melanin skin pigmentation. The right hand is of a teenager with iron overload caused by thalassaemia major. The left hand is of her mother, who has normal iron status.
46 / Chapter 4: Iron overload Cardiac damage due to iron is a dominant problem in transfusional iron overload. As for hereditary haemochroma- tosis, iron also damages the liver (Fig. 4.3) and the endocrine organs, including the hypothalamus and pituitary, with failure of growth, delayed or absent puberty, diabetes mellitus, hypothyroidism and hypoparathyroidism. Skin pigmentation as a result of excess melanin and haemosiderin gives a slate grey appearance even at an early stage of iron overload. In the absence of intensive iron chelation, death occurs in the second or third decade of life in thalassaemia major, usually from congestive heart failure or cardiac arrhythmias. T 2 * MRI is a valuable measure of cardiac and liver iron loading (Fig.4.4). Table 4.4 Causes of anaemia that may lead to transfusional iron overload. Congenital Acquired β-Thalassaemia major Myelodysplastic neoplasias β-Thalassaemia/Hb E disease Red cell aplasia Congenital dyserythropoietic anaemias Acute leukaemias Sickle cell anaemia (some cases) Aplastic anaemia Red cell aplasia (Diamond–Blackfan) Primary myelofibrosis Congenital sideroblastic anaemia Dyserythropoietic anaemia Hb, haemoglobin. (b) (a) Figure 4.3 β-Thalassaemia major: needle biopsy of liver. (a) Grade IV siderosis with iron deposition in the hepatic parenchymal cells, bile duct epithelium, macrophages and fibroblasts (Perls’ stain). (b) Reduction of iron excess in liver after intensive chelation therapy.
Chapter 4: Iron overload / 47 It can detect increased cardiac iron and predict for cardiac failure or arrhythmia before sensitive tests detect impaired cardiac func- tion. The shorter the relaxation time, the greater the cardiac iron burden and the greater risk of subsequent cardiac failure or arrhythmia (Fig. 4.5). Serum ferritin and liver iron correlate poorly with cardiac iron (Figs. 4.4 and 4.5). Moreover, serum ferritin is raised in viral hepatitis and other inflammatory disor- ders and should therefore be interpreted in conjunction with more accurate tests of iron status, such as T 2 * MRI, Ferriscan R2 or liver biopsy. It is, however, useful for monitoring changes in iron burden when this is being treated by chelation therapy. It is important to monitor LVEF annually by echocardiography or MRI from the age of 8 in thalassaemia major. Monitoring for liver, endocrine and bone disease is also needed. Iron chelation therapy Iron chelation therapy is used to treat transfusional iron overload. Three effective drugs are available: orally adminis- tered deferasirox and deferiprone, and parenterally administered 8000 6000 4000 Serum ferritin (μg/L) 2000 0 1000 3000 5000 7000 0 10 20 30 40 50 Heart T 2 * (ms) 60 70 80 90 100 Figure 4.5 Comparison of T 2 * magnetic resonance imaging (MRI) measurement of cardiac iron and serum ferritin in thalassaemia major patients. There are substantial numbers of patients with very high serum ferritin (>3000 μg/L) but normal heart iron (T 2 * >20 msecs) and, conversely, many patients with serum ferritin levels <1000 μg/L with severe cardiac iron loading (T 2 * <10 msecs). Source: L.J. Anderson et al. (2001) Eur. Heart J. 22: 2171–79. (b) (a) Liver Liver Liver Heart Heart Liver Spleen Spleen (d) (c) Figure 4.4 T 2 * magnetic resonance images (MRIs) showing tissue appearance in iron overload: (a) normal volunteer, (b) severe iron overload. Green arrow, normal appearance; red arrow, iron overload. Lack of correlation: liver and cardiac iron in two cases of thalassaemia major, (c) and (d).
48 / Chapter 4: Iron overload deferoxamine (Table 4.5). Thalassaemia major is the most fre- quent indication worldwide, but chelation is also used for iron overloaded, usually heavily transfused patients with the other anaemias (Table 4.4) and with other iron loading anaemias such as non-transfusion-dependent thalassaemia (Chapter 7) and pyruvate kinase deficiency (Chapter 6). Deferasirox is given orally once daily and leads to iron loss in the faeces. Skin rashes and transient changes in liver enzymes and rise in serum creatinine are the main side effects. Licensing for young children and lack of major side effects have resulted in its widespread use, but high cost means that the other drugs may be preferred in some countries. Deferasirox removes iron primarily from the liver and is the least effective of the three drugs for eliminating cardiac iron. Deferiprone is also an oral chelator and causes predomi- nantly urinary iron excretion. It was usually given in three doses daily but a twice daily delayed release formulation has now been approved. It is licensed for first line treatment at varying starting ages in children in different countries. It may be used alone or, if this is inadequate, in combination with deferoxamine infused on one or more days a week, since the drugs have an additive or even synergetic effect on iron excre- tion. Alone it is the most effective of the three drugs at remov- ing cardiac iron and improving left and right heart function. Side effects include an arthropathy, agranulocytosis (in about 1%), neutropenia, gastrointestinal disturbance and, rarely in patients with diabetes, zinc deficiency. Monitoring of the blood count weekly for the first 6 months, fortnightly for the next 6 months and then every 2–4 weeks or at time of blood transfusion is recommended for all patients receiving defer- iprone. Combination therapy with the two orally active chelators, if either alone is not sufficiently effective, has been effective without unexpected toxicity in several trials, but is not yet licensed in any country. The third drug, deferoxamine, was the first of the three drugs to be used in clinical practice. It is not active orally and is usually given by subcutaneous infusion over 8–12 hours for 5–7 days each week; vitamin C is given to increase iron excre- tion. Most iron is lost in the urine, but up to one-third is also excreted in the stools. Because of the difficult administration, lack of patient adherence is a major problem. It may be given on one or more each days each week in combination with daily deferiprone or deferasirox and can be used intravenously in combination with oral deferiprone in patients with severe iron overload at risk of dying from cardiac failure. Side effects are particularly frequent if high doses are used in children and in adults without heavy iron overload. These include high tone deafness, retinal damage, bone abnormalities and growth retar- dation. Patients receiving deferoxamine should have auditory and fundoscopic examinations annually. All three chelators can be given in children. Deferasirox is most frequently used and a liquid formulation of deferiprone and a sprinkle form of deferasirox are available. Chelation is typically started in thalassaemia major after 10–12 units have been transfused or the serum ferritin is >1000 μg/L. In other conditions such as myelodysplastic neo- plasias, there is controversy about when to initiate chelation and hepatic and cardiac T 2 * MRI may help guide this decision. Table 4.5 Characteristics of desferrioxamine, deferiprone and deferasirox. Desferrioxamine (DFO) Deferiprone (DFP) Deferasirox (DFX) Structure Hexadentate Bidentate Tridentate Molecular weight (Da) 560 139 373 Iron–chelator complex 1 : 1 1 : 3 1 : 2 Plasma clearance (t 1/2 ) 20 min 1–3 h 1–16 h Absorption Negligible Peak 45 min Peak 1–2.9 h Iron excretion Urine + faecal Urine Faecal Therapeutic daily dose 40 mg/kg 75–100 mg/kg 20–40 mg/kg (dispersible tablet) 7–21 mg/kg (coated tablet) Route Parenteral Oral Oral Clinical experience >45 y >35 y >20 y Side effects Ototoxicity, retinal toxicity, growth defects, cartilage and bone abnormalities Agranulocytosis, arthropathy, gastrointestinal disturbance, transient transaminitis, zinc deficiency Skin rashes, gastrointestinal disturbance, rising serum creatinine
Chapter 4: Iron overload / 49 Chelation is given to keep the cardiac T 2 * at >20 msecs, liver at <7 mg/g dry weight and serum ferritin level at <1000– 1500 μg/L, when the body iron stores are approximately 5–10 times normal. MRI assesses cardiac and liver iron accurately and should be repeated annually or more frequently if there is definite cardiac or liver damage (Fig. 4.5). Serum ferritin is useful in monitoring changes in iron stores, but as it is an acute phase reactant it may be elevated in the presence of recent infection trauma or surgery and this may falsely suggest inad- equate chelation. Serial tests of heart, liver and endocrine function are also needed to monitor therapy. Life expectancy has improved dramatically for thalassaemia major patients since the introduction of iron chelation. Chelation may even reverse liver, endocrine and cardiac damage in cases where this has developed before effective chelation has been started. ■ Iron overload may be caused by excessive absorption of iron from food because of low plasma hepdidin levels due to hereditary (genetic) haemochromatosis or because of raised plasma erythroferrone levels due to ineffective erythropoiesis. ■ Iron overload may also result from repeated blood transfusions in patients with refractory anaemias. Each unit of transfused blood contains 200–250 mg of iron. ■ Excess iron absorbed from the gastrointestinal tract in hereditary haemochromatosis accumulates in the parenchymal cells of the liver, the endocrine organs and, in severe cases, the heart. ■ Hereditary haemochromatosis is usually caused by homozygous mutation C282Y of the HFE gene resulting in the HFE protein change and a low serum hepcidin level. Rarer forms are caused by mutations of other genes coding for proteins hemojuvelin, hepcidin, transferrin receptor 2 and ferroportin involved in iron regulation. ■ Repeated venesections are used to reduce the body iron burden in hereditary haemochromatosis. ■ Transfusional iron overload most frequently occurs in thalassaemia major, but also in other transfusion- dependent refractory anaemias, e.g. some cases of myelodysplastic neoplasias, sickle cell anaemia, primary myelofibrosis, red cell aplasia and aplastic anaemia. ■ Transfusional iron overload causes damage to the liver, endocrine organs and heart, with iron accumulation also in macrophages of the reticuloendothelial system. ■ Cardiac failure or arrhythmia caused by cardiac siderosis, best detected by T 2 * MRI, is the most frequent cause of death from transfusional iron overload. ■ Treatment is with iron chelating drugs: deferasirox and deferiprone which are active orally, or with deferoxamine, given subcutaneously or intravenously. ■ Life expectancy has improved dramatically in thalassaemia major as a result of iron chelation therapy and the use of T 2 * MRI to accurately measure cardiac and liver iron. SUMMARY Now visit www.wiley.com/go/haematology9e to test yourself on this chapter.
Chapter 5: Megaloblastic anaemias and other macrocytic anaemias / 51 Co + CH 3 Nucleotide Figure 5.1 The structure of methylcobalamin, the main form of vitamin B 12 in human plasma. Other forms include deoxyadenosylcobalamin, the main form in human tissues; hydroxocobalamin and cyanocobalamin, the main forms used in treatment of vitamin B 12 deficiency and in multivitamin supplements. Introduction to macrocytic anaemia In macrocytic anaemia, the red cells are abnormally large (mean corpuscular volume, MCV >98 fL). There are several causes (Table 2.5) broadly subdivided into megaloblastic and non-megaloblastic (Table 5.10), based on the appearance of developing erythroblasts in the bone marrow. An elevated MCV may also be an artefact reported by an automated cell counter if red cell agglutination or a paraprotein is present. Megaloblastic anaemias This is a group of anaemias in which the erythroblasts in the bone marrow show a characteristic abnormality: matu- ration of the nucleus is delayed relative to that of the cytoplasm. The underlying defect is defective DNA synthe- sis. This is usually caused by deficiency of vitamin B 12 or folate. Less commonly, inherited or acquired, e.g. by drugs, abnormalities of the metabolism of these vitamins or inherited or acquired lesions in DNA synthesis may cause an identical haematological appearance (Table 5.1). Morphologically, this developmental asynchrony is manifest by a persistently open, loosely organized chromatin in the erythropoietic cell nucleus, while the cytoplasm exhibits staining changes of haemoglobi- nization typical of later stages of maturation (Fig 5.14). Vitamin B 12 (B 12 , cobalamin) Vitamin B 12 is synthesized in nature by microorganisms. Animals acquire it by eating food of animal origin, by internal production from intestinal bacteria (not in humans) or by eat- ing bacterially contaminated foods. Vitamin B 12 consists of a small group of compounds, the cobalamins, which have the same basic structure, a cobalt atom at the centre of a corrin ring (Fig. 5.1). The reactive centre of the molecule is attached to either a cyano group (-CN, cyanocobalamin), a hydroxyl group (-OH, hydroxocobalamin), a methyl group (-CH 3 , methylco- balamin) or 5-deoxyadenosyl (adenosylcobalamin). The vita- min is found in foods of animal origin such as liver, meat, fish and dairy produce, but does not occur in fruit, cereals or veg- etables, except in small amounts due to contamination by insect parts in harvesting or by micro-organisms in a natural environment (Table 5.2). Absorption A normal diet contains a large excess of B 12 compared with daily needs (Table 5.2). B 12 is released from protein binding in food by pepsin in the stomach. It is then mainly combined with the protein, intrinsic factor (IF). IF is synthesized by the gastric parietal cells. The IF–B 12 complex subsequently binds in the ileum to a specific surface receptor for IF, cubam, a complex of proteins cubilin and amnionless. Amnionless directs endocytosis of the cubilin IF–B 12 complex into the ileal cell so that B 12 is absorbed and IF destroyed (Fig. 5.2). The maximum amount of B 12 that can be absorbed from a single oral dose (either in the form of food or a supplement) via the IF–cubam mechanism is about 1–2 μg. Some dietary B 12 , after release from food, binds to the gly- coprotein haptocorrin (also known as R-factor, R-protein, or transcobalamin I) (Fig 5.2). This glycoprotein is present in plasma, milk, saliva and gastric juice. Release of dietary B 12 from haptocorrin, making it available for binding to IF, depends largely on proteases from the pancreas. Table 5.1 Causes of megaloblastic anaemia. Vitamin B 12 deficiency (causes are listed in Table 5.3) Folate deficiency (causes are listed in Table 5.5) Combined folate and B 12 deficiency Abnormalities of vitamin B 12 or folate metabolism, e.g. transcobalamin deficiency, nitrous oxide, anti-folate drugs such as methotrexate, phenytoin or trimethoprim Inherited defects of DNA synthesis Congenital enzyme deficiencies, e.g. orotic aciduria, which impairs pyrimidine synthesis Acquired enzyme deficiencies, e.g. due to hydroxyurea, purine synthesis antagonists such as 6-mercaptopurine, pyrimidine antagonists such as cytarabine
52 / Chapter 5: Megaloblastic anaemias and other macrocytic anaemias Table 5.2 Vitamin B 12 and folate: nutritional aspects. Vitamin B 12 Folate Typical daily dietary intake 7–30 μg 200–400 μg Food sources Animal products only Many foods, especially liver, greens and yeast Effect of cooking Little effect Easily destroyed Minimal adult daily requirement 2 μg 100–200 μg Body stores when replete 2–3 mg (sufficient for 2–4 years) 10–12 mg (sufficient for 4 months) Absorption Site Mechanism Limit Ileum Bound to intrinsic factor, absorbed by cubam 2–3 μg/day Duodenum and jejunum Conversion to methyltetrahydrofolate 50–80% of dietary content Enterohepatic circulation 5–10 μg/day 90 μg/day Transport in plasma Most bound to haptocorrin; TC essential for cell uptake Weakly bound to albumin Major intracellular physiological forms Methyl- and deoxyadenosyl-cobalamin Reduced polyglutamate derivatives Usual therapeutic form Hydroxocobalamin or cyanocobalamin Folic (pteroylglutamic) acid TC, transcobalamin (transcobalamin II); haptocorrin = transcobalamin 1. Bile Portal plasma Liver, bone marrow, other cells Diet Food–B12 HC–B12 HC–B12 HC–B12 Enterohepatic circulation Pancreatic trypsin Ileum IF receptor (cubulin/ amnionless) TC–B12 TC receptors (CD320) TC degraded B12 methyl B12 ado B12 Parietal cell IF IF–B12 HC Stomach IF-B12 Duodenum and jejunum Figure 5.2 The absorption of dietary vitamin B 12 after combination with intrinsic factor (IF), through the ileum. TC, transcobalamin; IF, intrinsic factor; HC, haptocorrin; ado B 12 , see text.
Chapter 5: Megaloblastic anaemias and other macrocytic anaemias / 53 Transport of vitamin B 12 : the transcobalamins Vitamin B 12 is absorbed from the ileal cell into portal blood, where it attaches to the plasma-binding protein transcobala- min (TC, also called transcobalamin II), which delivers B 12 to the bone marrow and all other tissues. Although TC is the essential plasma protein for transferring B 12 into the cells of the body, the amount of B 12 on TC is normally very low (<50 ng/L). Congenital TC deficiency due to germline mutations in the TCN2 gene causes megaloblastic anaemia because of failure of B 12 to enter marrow (and other cells) from plasma, but the serum B 12 level in TC deficiency is normal. This is because most B 12 in plasma is bound to haptocorrin. This glycoprotein is synthesized by granulocytes and macrophages. In myelopro- liferative neoplasms where granulocyte production may be greatly increased, the haptocorrin and B 12 levels in plasma both rise considerably. This also occurs in some liver diseases. B 12 bound to haptocorrin in the blood does not transfer to mar- row; the B12 is functionally ‘dead’. Biochemical functions Vitamin B 12 is a coenzyme for two biochemical reactions. First, as methyl B 12 it is a cofactor for methionine synthase, the enzyme responsible for methylation of homocysteine to methionine which uses methyltetrahydrofolate (methylTHF) as methyl donor (Fig. 5.3). Second, as deoxyadenosyl B 12 (ado B 12 ) it is coenzyme in the conversion of methylmalonyl coen- zyme A (CoA) to succinyl CoA, a key intermediate in the citric acid cycle (Fig. 5.3). Folate Folic (pteroylglutamic) acid is the parent compound of a large group of compounds, the folates (Fig. 5.4). This family of compounds is also known as vitamin B9, a name which often appears on the packets of cereals that have been fortified with the vitamin. Absorption, transport and function Dietary folates are a complex mixture of variously reduced and polyglutamated folates. They are all converted to one com- pound, methylTHF, a reduced monoglutamate form which circulates in plasma (Fig. 5.5a). After entering cells, methyl- THF is de-methylated to THF, and then converted to folate polyglutamate forms by addition of usually four, five or six glutamate moieties (Fig. 5.6). Folic (pteroylglutamic) acid itself is a poor substrate for reduction by dihydrofolate reductase. It is mainly converted during absorption to methylTHF at doses of 200–400 μg, larger oral doses of folic acid enter portal plasma unchanged and are then converted to physiological forms in the liver or excreted in the urine (Fig 5.5b). Folates are needed in a variety of biochemical reactions in the body involving single carbon unit transfer, in amino acid interconversions, e.g. homocysteine conversion to methionine (Figs. 5.3 and 5.6) and serine to glycine, or in synthesis of pyrimidine and purine precursors of DNA. Biochemical basis for megaloblastic anaemia DNA is formed by polymerization of the four deoxyribonu- cleoside monophosphates derived from their triphosphates (Fig 5.6). Folate deficiency is thought to cause megaloblastic anaemia by limiting synthesis of thymidine monophosphate (dTMP) from deoxyuridine monophosphate, a rate-limiting step in DNA synthesis. This reaction needs 5,10-methylene THF polyglutamate as coenzyme. Consequent starvation of the precursor dTTP leads to prolongation of the S phase dur- ing mitosis, failure to form new double-stranded DNA and apoptotic cell death. The role of B 12 in DNA synthesis is indirect. B 12 is needed in the conversion of methylTHF, which enters marrow and other cells from plasma, to THF and other active folates. In this reaction, homocysteine is converted to methionine. THF (but not methylTHF) is a substrate for folate polyglutamate synthesis. The folate polyglutamates are the intracellular folate coenzymes. B 12 deficiency therefore reduces the supply of the critical folate coenzyme 5,10-methylene THF polygluta- mate, needed for synthesis of thymidine monophosphate (dTMP) (Fig. 5.6). The block caused by B 12 deficiency in the conversion of methylTHF either mono- or poly-glutamate, to other forms of folate has been termed the ‘methylfolate trap’. Other congenital or acquired causes of megaloblastic anaemia, e.g. antimetabolite drug therapy, mainly inhibit purine or pyrimidine synthesis at one or another step. The result is a reduced supply of one or other of the four precursors needed for DNA synthesis. Folate reduction During the synthesis of dTMP, the folate polyglutamate coen- zyme becomes oxidized from the THF to the dihydrofolate (DHF) state (Fig. 5.6). The enzyme dihydrofolate reductase regenerates active THF from DHF. Inhibitors of this enzyme, Homocysteine S-adenosyl homocysteine Methylation of DNA, myelin, amines, proteins, etc. Propionyl CoA Methylmalonyl CoA Succinyl CoA S-adenosyl methionine Methionine Methyl THF THF Methyl B12 Ado B12 Figure 5.3 The biochemical reactions of vitamin B 12 in humans. Ado B 12 , deoxyadenosylcobalamin; CoA, coenzyme A; THF, tetrahydrofolate.
54 / Chapter 5: Megaloblastic anaemias and other macrocytic anaemias e.g. methotrexate, inhibit folate-mediated biochemical reactions including in DNA synthesis (Fig. 5.6). Methotrexate is a useful drug, mainly in the treatment of malignant, e.g. acute lymphoblastic leukaemia (Chapter 17) or inflammatory disease, e.g. rheumatoid arthritis, psoriasis with excessive cell turnover. The weaker antagonist, pyrimethamine, is used pri- marily against toxoplasmosis. Trimethoprim, active against bacterial DHF reductase but only very weakly against the human enzyme, is used alone or in combination with a sul- phonamide, as cotrimoxazole, especially to treat urinary tract infections. Toxicity caused by methotrexate or pyrimethamine may be reversed by the reduced folate, folinic acid (5-formyl THF). In the protocols used, this reversal does not eliminate the effectiveness of methotrexate when used in anti-cancer chemotherapy. Causes of severe vitamin B 12 deficiency In developed countries, severe deficiency is usually caused by the auto-immune disease, pernicious anaemia (Table5.3). Less commonly, severe deficiency may be caused by extreme lack of B 12 in the diet (as in strict veganism without dietary supplements), total gastrectomy or small intestinal lesions. In vegetarians and people subsisting on a poor-quality diet low in B 12 -rich foods, an intact entero-hepatic circulation usually but not invariably helps to protect them from severe B 12 deficiency. The deficiency takes at least 2 years to develop, i.e. the time needed for body stores to deplete at the rate of 1–2 μg/day when there is severe malabsorption of B 12 . There is no syndrome of B 12 deficiency as a result of increased utiliza- tion or loss of the vitamin. Nitrous oxide, however, may rapidly inactivate body B 12 (p. 63). Pernicious anaemia Pernicious anaemia (PA) is caused by autoimmune attack on the gastric mucosa, leading to atrophy of the stomach. The wall of the stomach becomes thin, with a plasma cell and lymphoid infiltrate of the lamina propria. Intestinal metaplasia may occur. Destruction of parietal cells results in achlorhydria and lack of secretion of IF. Serum gastrin levels are raised. Helicobacter pylori infection may initiate an autoimmune 1 2 8 7 6 9 10 5 N N CH 2 N H N H N H N C C CH CH 2 CH 2 O O O – O – O H 2 N 3 4 C O 1 2 8 7 6 9 10 5 N NH CH 2 HN H N H N H N C C CH CH 2 CH 2 O O O – O H 2 N 3 4 CH 3 One carbon substituent Tetrahydrofolylpoly-γ-Glutamate (H 4 PteGlu n ) Folic acid, Pteroylglutamate (PteGlu) Pteroic acid Pteridine p-Aminobenzoic acid Methyl Methylene Methenyl Formyl Formimino N-5 Methanol N-5, N-10 Formaldehyde N-5, N-10 Formate N-5 or N-10 Formate N-5 Formate Position Oxidation state CHO CH 2 CH CH HN C O n–1 H N C CH CH 2 CH 2 O O – O – C O Glutamate Figure 5.4 The upper panel shows the chemical structure of folic acid (pteroylglutamic acid). The middle panel shows the structure of folic acid reduced to the tetrahydrofolate form and with one additional glutamic acid. The lower panel shows the five single carbon units that may be added to the folate molecule. Source: Courtesy of Professor Barry Shane.